rrl BRITL 543.1.B62 c. 1 BLYTH # FOODS iiiiiiiiiimi r. 3 T1S3 00133237 ■$ OS (!) 5 O < > This Book may be kept out FOODS: THEIR COMPOSITION AND ANALYSIS. LEAF LARD x185. FAIRBANK'S O L E OS T E A R I N E x65. MICROSCOPICAL CHARACTERS OF LARD AND BEEF CRYSTALS. . 6 FOODS: THEIE COMPOSITION AND ANALYSIS. A MANUAL FOR THE USE OF ANALYTICAL CHEMISTS AND OTHERS. WITH AN INTRODUCTORY ESSAY ON THE HISTORY OF ADULTERATION. ALEXANDEE WYNTER BLYTH. M.K.O.S., F.I.C., F.C.S., &C., BARRISTER AT LAW ; PCBLIC ANALYST FOR THE COUNTY OF DEVON, AND MEDICAL OFFICER. OF HEALTH AND PUBLIC ANALYST FOR ST. MARYLEBONE. With Numerous Tables and Illustrations. FOURTH EDITION, REVISED AND ENLARGED. LONDON: CHARLES GRIFFIN AND COMPANY, LTD. EXETER STREET, STRAND. NEW YORK: D. VAN NOSTRAND COMPANY, 23 MUHRAY STREET AND 27 WARREN STREET. 1896. 1 PEEFACE TO THE FOURTH EDITION. The progress of analytical chemistry as applied to the investiga- tion of Foods has made such rapid strides since the previous edition of the present work, that the author has had consider- able difficulty in revising the pages so as to embody all that is of value, and, at the same time, retain the handy size of the older edition. Processes which experience has shown to be faulty or imper- fect, have been omitted and replaced by new matter, and greater attention has been paid to the application of purely physical methods. A number of new diagrams and new tables have been added. The author believes that the work in its present form not only embodies the results of his own personal experience, but fairly represents the views and details the methods of the modem Food-analyst. The Court House, St. JlAKYLEBONE, W., February, 1896. PREFACE TO THE SECOND EDITION. The following pages are an instalment of the Second Editiou of the Author's " Manual of Practical Chemistry " — the First Part of the New Edition being now issued separately under the title of "Foods: their Composition and Analysis," and the Second Part under that of " Poisons : their Effects and Detection." The reasons for the alteration of the title are sufficiently ob- vious : the present appellation is distinctive, rendering impos- sible any confusion between this Manual and others (of a widely different scope and manner of treatment) which might come equally under the designation, " Practical Chemistry." The present Volume, however, is not a mere repiint of the Division, " Foods" in the First Edition. It has been thoroughly revised and re-wiitten, where necessary, and enlarged by the addition of new matter to more than double the number of pages allotted to the subject in the original work. The Historical Introduction prefixed is the result of consider- able labour and research, and, it is hoped, will be found — together with the review of English Legislation, Past and Present, relative to Adulteration — not without interest. As in the First Edition, abstracts of a few legal cases are given at the end of the chief Articles. These have been carefully seleeted, as illustrative either of ingenious defence, or of certain points in the Adulteration Acts. It has often been remarked that private individuals rarely avail themselves of the "Sale of Food and Drugs " Act. This, probably, is due to insufficient acquaintance with the technical details of the mode of proce- dure, and the Author has, therefore, been careful to explain the " Purchase " sections fully in their relations both to the official Inspector and to the private purchaser. In the Appendix will be found the Text, entire, of the English laws at present in operation, as well as the best and most recent of the American Acts relating to the Adulteration of Food. In the Scientific Portion of the work, the professional Chemist will find details of most of the processes of any value in Food Analysis hitherto published, and in all cases (either by the aid of Footnotes, or in the Bibliography appended to each Article) the original source of the information is indicated. In addition, ai-e given a large number of Processes, either invented or im- proved by the Author, and not previously published — such, e.g., as those desciibed in the Articles on Milk, Butter, Tea, Flour, Water, &c. Numerous Tables, some of which are indispensable and others convenient, have also been added ; and new Illustrations, from original drawings, introduced. The Article on Milk — a special feature of the First Edition — is still further enlarged, and contains the Author's most recent researches on the subject. It may, perhaps, be considered a fairly complete Monograph. In the Article on Water (added by request) the application of an improved jDrocess for combustion in. a vacuum is detailed, and the importance of Biological methods of examination is insisted upon — not as supplementary to Chemical tests, but as of equal (if not of superior) value to these. Though the scope of the Manual is mainly that of a Labora- tory Handbook, yet the dietetic and medical aspects of the more important Foods are, where necessary, fully considered, and the Author believes that a great proportion of the work is thus of that general interest which will render it useful to those who, without much chemical knowledge, yet desire to have, in a form admitting of easy reference, the latest information relative to Foods and Beverages. In conclusion, he can only express a hope that the work, in its new shape, will be found widely useful, and more wox'thy of the very kind reception accorded to the First Edition. Court House, St. Maetleboiti, April, 1883. CONTENTS. PART I.— HISTORY OF ADULTERATION. I. EARLY NOTICES OF ADULTERATION, ESPECIALLY IN ENGLAND, Section 1. Roman and Greek Notices of Adulteration, .... 2. Low Standard of Commercial Morality in the Eleventh and Twelfth Centuries, ........ Assizes of Bread— Bakers. 3. Early Regulations in England relative to the Assize of Bread, . Statute of Assize, 1582, Frauds of the Bakers Brewers and Vintners. 4. Beer, Fraudulent Practices of early Brewers, Ale-tasters, . 5. Wine, ........... Spices— Drugs. 6. Regulations of the Pepperers, Regulations of the Druggists and Grocers, .... Tago 3 II. adulteration in FRANCE. 7. General Inspection of Provisions in Olden Time by the "Police des Commissaires," and Ancient Regulations relative to the Adulteration of Beer, . S. Flour and Bread — Various Decrees relative to, Punishments of Freuch Bakers by Penance, &c., 9. Wine — A Curious Decree of the Pi'ovost of Paris. Inspectors of Wines and Drinks appointed. Poisonous W^ine, ...... " Vin de Raisin de Bois," .... 10. Butter — Regulations relative to the Sale of, 11. Drugs 12. Conseils de Salubrite', X CONTENTS. III. ADULTERATION IN GERMANY. Section Page 13. The Old German Guilds, 15 14. Old German Eegulatioiis relating to Bread, .... 16 15. ,, ,, Wine 16 10. ,, ,, Drugs— Establishment of the Continental Pharmacopcfiias, ..... 13 rV. HISTORY OF ENGLISH LEGISLATION WITH REGARD TO THE ADULTERATION OF FOOD. 17. First General Food Act, 1S60 18 18. Bakers— Bread, 19 19. Acts relating to Beer and Porter, 20 20. „ „ Wines, 20 21. The Tea Acts, 21 22. The Coffee Acts— History of E.egulations as to Chicory and Coffee, 21-24 23. The Select Committee, 1S55, 24 The Analytical Sanitary Commission of the "Lancet," . . 25 24. The Adulteration Acts, 1860 and 1872, 26-28 25. The Select Connnittee, 1 874, 28 26. The Sale of Food and Drugs Acts, 1875 and 1879, ... 29 The "Prejudice" question, ....... 29 Sale of Food and Drugs Act, 1879, 31 V. HISTORY OF THE PRESENT SCIENTIFIC PROCESSES FOR THE DETECTION OF ADULTERATION. 27. Early Workers 32 28. Discovery of Milk-sugar by Bartoletus, ..... 32 The Experiments of Francesco Redi, &c., .... 32 29. The Works of the Hon. Robert Boyle, and of J. B. Vandeu Sande, 33 30. The Invention of the Microscope 34 Antony Van Leeuwenhoek, Dr. Hy. Power, Ehrenberg, Donne, 34-37 31. General Advance of Chemistry : i^eumann tJaspar, Boerhaave, Berzelius, &c., ........ 37 32. Accum's work. — " Death in the Pot, " 37 33. The Works on Adulteration of Food, &c., by Bussy aud Boutron- Charlard, Gamier and Harel, and Friedrich, ... 40 34. The Works of John Mitchell, Chevallier, and Normandy, . 41 35. Chevallier's Dictionary — Alphonse Normandy's Handbook, . 41 36. Dr. Hassall's Contributions to the " Lancet," .... 42 37. The Establishment of the Society of Public Analysts, . . 42 The "Analyst" — "Limits," ....... 43 38. A list of General Treatises on Adulteration chronologically arranged, .......... 43 VI. THE PRESENT LAW IN ENGLAND RELATIVE TO ADULTERATION OF FOOD. The Sale of Food and Drugs Act, 38 and 39 Vict., c. 63, and Sale of Food and Drugs Act Amendment, 42 and 43 Vict., c. 30. 39. Preamble of the Act ; Definition of "Food and Drugs," . . 46 The Mixing, Colouring, or Staining of Foods, and the Com- pounding of Drugs, . 47 Section Paso 40. The "Prejudice" question io the Acts, 48 Regulations or Limits as to the Strength of Spirits, . . 49 Drugs to be sold only in Accordauce with the Demand of the Purchaser— No Offence if there is a Label Distinctly de- scribing the Article sold — Abstraction from any Article of Food of any Constituent likely to impair its properties, &c., 49 4L The Label Section— Appeal Case of Liddiard v. Reece, . . 49 The Question of Completeness of Sale, 50 42. Question as to how far Notices over Shop Doors, &c., protect a Vendor, 51 43. Appointment and Qualifications of Analysts, .... 52 44. The Purchase of Samples by a Purchaser for Analysis, . . 54 45. The Procuring of Samples for Analysis by Medical Officers of Health, &c., 55 Penalty for Refusal to Sell -The Case of Rouch v. Hall, . . 55 46. Method to be pursued by a Purchaser under the Act, . . 56 47. Regulations of the General Post Uthce relative to the Trans- mission of Samples through the Post, .... 58 48. The Certificate of the Analyst — The Institution of Proceediags — Quarterly Reports of the Analyst— The Certificate of the Analyst is Evidence, 59, 00 49. Provision for Analysis of the Sample at Somerset House— The Defendant may Prove by Written Warranty that he had no Reason to believe the Article sold was any other than Pure, &c., 60 Provision for Payment of Penalties — The forging of Warranties — The giving of False Labels — The Proceeding by Indict- ment—The Examination of Tea on Importation, . . 61 VII. THE DUTY OF THE INSPECTOR OR PURCHASER UNDER THE ACT- SO. Giving full details as to the Duties of an Inspector uuder the Act, 62-64 PART XL— INTRODUCTORY. I. DESCRIPTION OF A FEW SPECIAL FORMS OF APPARATUS USEFUL IN FOOD ANALYSIS. 51. Soxhlet's Fat-extracting Apparatus with Various iSIodifications, 67 Wynter Blyth's Ether Tube and " Ether Recovery" Apparatus, 68, 69 52. The Spiral Balance, . 70 53. Vacuum Processes — The Mercury Pump, .... 71 II. THE MICROSCOPE, THE SPECTROSCOPE, AND THE ART OF PHOTOGRAPHY AS APPLIED TO THE CHEMISTRY OF FOOD. 54. The best Form of INIicroscope— Manipulation of Tissues, . . 72-74 55. The Micro-spectroscope, 74 The Sorbv-Browninc.- Micro-spectroscope — jSIeasurement of Bands, . .° ....... . 74-78 The Hema-spectroscope of M. de Thierry, .... 78-79 Xll CONTENTS. Section Page 56. Spark Spectra — The method of L. de Boisbaudran, . . 79, 80 56a. The Polarising Colorimeter — The Colorimeter with the Lumner-Brodhuns Prism, 80-84 56&. Quantitative Spectroscopy — Karl Vierordt's divided Slit — The Extinction Coefficient — The Absorption Spectrum of Permanganate, ........ 84-89 57. Photography — Stein's Photographic Microscope — Dr. Wood- ward's Method, 90 58. Colour — Mr. Sorby's Method of Examining Colouring-Matters, 91, 92 59. Red Colouring-Matters — Cochineal — Absorption Factors for C'ar- minic Acid — Aniline Keds — Absorption Factors for Fuch- sine — Safranine — Coralline — Aurine — Ponceau — Congo Eed — Erythrosine — Fast Red— Biebrich Scarlet— Cioceine Scarlet — Alkanet — Madder — Alizarine — Purpurine — Safflower — Logwood — Effect of Alum on Logwood Spec- trum — Brasiline — Santalin — The Red Colouring- Matters of Fruits and Berries, 92-100 60. Orange and Yellow Colouring-Matters — The Annatto Colours — Turmeric — Picric Acid — Fustic — Chrysophanic Acid — Gamboge — Aniline Oranges and Yellows — Croceine Orange — Resorcin Yellow — Diphenylamine Yellow — Diphenylamine Orange — Chrysoidine — Methyl Orange — Metanil Yellow— Phosphine, 100-103 6L Green Colouring- Matters — Chlorophyll Group of Colours — Aldehyde Green— Iodine Green— Malachite Green, . . 104-106 62. Indigos and Violets — Indigo — Indigo tin — Litmus — Aniline Blues — Methylene Blue — Aniline Blue— Methyl Violet — Alkali Blue, . . . . _ 106-108 63. Brown Colours — Bismarck Brown — Acid Brown — Caramel, . 108, 109 64. Scheme for the Detection of the Aniline and other Colouring- Matters by Chemical Reagents, ..... 109-113 64a. Witt and Weingartner's Classification of Aniline Colours, - 113-117 The Mineeal Matters or "Ash" of Food. Analysis of the Ash of Organic Substances. 65, The Percentage of Ash— Ash Soluble in Water — Ash Soluble in Acid — Alkalinity of Ash — Percentage of Chlorine — Phosphoric Acid in the Ash, 117-119 66. General Method of Determining all the Constituents in an Ash — Determination of Carbon Dioxide — Of Sulphuric and Phosphoric Acids — Bunge's Process for the Alkalies — Laugier's Method of Determining the Ash of Sugar, . 117-123 66a. Methods of Estimating Nitrogen and Nitrogenous Substances in Foods— Estimation of Total Nitrogen by Kjeldahl's Method — Estimation of Nitrogen from Albximen — Esti- mation of non-Albuminoid Nitrogen — Estimation of Am- monia — Estimation of Amido-Acid and Amide Nitrogen, 123-125 CONTEXTS. xiii PART III. -CARBO-HYDRATES. Starchy and Saccharine Substances. Section Page 67. Classification of the Carbohydrates — I. The Grape Sugar Group— II. The Cane Sugar Group — III. The Carbo- hydrates—Cane Sugar — Compounds of Inorganic Bases with Sugar, 126-129 68. Adulterations of Sugar — The Detection of Dextrin — Of Starch Sugar— Of Mineral Matters, 129-132 69. Full Analysis of Sugar — Methods of Taking the Ash of Sugar — Laugier's Method, 1,32-1.35 70. Glucose or Grape Sugar— How to obtain it — Properties, .135,136 71. Levulose, 136 Estimation of Sugar. 72. Estimation of Sugar — 1. Chemical Processes depending on the Precipitation of Copper Suboxide — 2. Volumetric Processes by Aid of Solutions of Salts of Mercury — Knapp's Solution— Sacchse's Solution, 137, 13S 73. Soxhlet's Researches on the Behaviour of Invert, Milk-Sugar, Galactose, and Maltose to Fehling's Solution — The Cyanide Copper Process — Dr. Pavy's Process— Physical Processes for the Determination of Sugar — Mitsclierhch's Polariscope— Soleil's Saccharimeter — Jellett's and other Polarimeters, 139-150 Confectionery— Sweetmeats. 74. Flavouring and Colouring-Matters of Sweetmeats — Composi- tion of Sweetmeats, ........ 150, 151 75. Analysis of Sweetmeats, 151, 152 Honey. 76. General Description of Honey— Analyses of Honey — Characters of Pure Houey — Adulterations of Honey — Artificial Honey, 153-157 Treacle — Molasses. 77. Composition and Analysis of Treacle, 157, 158 .Jam and Preserved Fruits. 78. General Composition of Jam, ....... 79. Microscopical Structure of Apples, Pears, Damsons, Plums Oranges, and Lemons, the Strawberry, Raspberry, Goose berry, Blackberry, and the Currant, Saccharin, 159-162 163 Starch. 80. Starch, its Composition — Method of Estimation, . . . 163-169 Microscopical Identification of Starches, . . . . 170, 171 Division of Starches, 171-176 XIV CONTENTS. Section Page 81. Vogel's Division of the Starches, 176,177 Karmarsch and Wiesner's Values, ...... 177, 178 Bibliography relative to the Starches, . , . . 178 Wheat — Wheaten Flour. 82. Varieties and Composition of Wheat, . . . . . 179 S3. Constituents of Flour, 179-186 Analysis of Flour. 84. Mici'oscopical Examination of Flour, . . . . . 186 ,, Structure of Corn-cockle, 187 Detection of Ergot in Flour, 189 Potato Starch, 190 ,, Leguminous Starches in Flour, . . . . 191 85. ,, Alum and Mineral Matters generally in Flour, . 193-196 86. Proximate Analysis of Flour, 196-198 87. Legal case relative to Flour, 198 Bread. 88. Definition of Bread— The Process of Making Bread, . . 199 General Composition of Bread, 199,200 Alteration of Bread by Moulds. &c., 201,202 89. Adulterations of Bread 202 90. Alum in Bread — Its Influence on Health : Methods for its Estimation, 203-207 Bibliography relative to Flour and Bread, .... 208, 209 91. Infants' Farinaceous Foods, 209, 210 Oats — Oatmeal. 92. Composition of the Oat — Oatmeal — Adulteration of Oatmeal, 210, 211 Barley. 93. Barley— Barley Bread, 212, 213 Rye. 94. Composition of Rye Flour— Rye Bread, 213, 214 Rice. 95. Composition of Rice— The Ash of Rice, 214, 215 Maize. 96. Composition of Maize— Ash of Maize, 216,217 Millet. 97. Composition of Millet— Ash of Millet, 217 Potato. 98. Composition of the Potato — The Fungus Producing the PotatoDisease— Analysis of the Potato,. . . . 218-221 Peas. Section Pajre 99. General Composition of Peas — Analysis of Peas, . . . 222-224 100. Preserved Peas — Copper in Peas — Discussion as to Poisonous Effects of Coppered Peas, 224, 227 Chinese Peas. 101. Composition of Chinese Peas, 227 Lentils. 102. Composition of Lentils, 228 Beans. 103. Composition of the Kidney and the Broad Bean, . . , 228, 229 PART IV.— MILK, CREAM, BUTTER, CHEESE. MILK. Historical Introduction. 104. Early Ideas as to the Composition of Milk : Aristotle— Avi- cenna — Placitus — Panthaleon— Bartoletus — Bartholomew- Martin — Ludovico Testi — Leeuwenhoek, . . . 230, 231 105. Boerhaave's Views on Milk, 231-233 106. Early Quantitative Analyses of Geoffrey and Doorschodt, . 233, 234 107. The Experiments of VuUyanoz as to Milk-Sugar, . . . 234 108. The Investigations of Voltelenus on Milk, .... 234, 235 109. The Experiments of Schoepff, Scheele, Hoffman, and Caspar Neumann, . 235, 23(> The Composition of Cow's Milk. 110. The General Physical Properties of Milk, .... 236-238 111. The Amphioteric Reaction of Milk, 2.38 112. Total Solids of Milk 238,239 113. Milk-Fat 239 114. Chemical and Physical Properties of Palmitin, Stearin, Olein, and other Constituents of Milk-Fat 239-242 115. The Albuminoids of Milk, 242-215 116. Milk-Sugar, 245, 246 117. Mineral Constituents of Milk, 246,247 118. Other Constituents of Milk — Lacto-Proteine, Galactin, Lacto-chrome 247, 248 119. Bitter Principles in Milk— Kreatinine, 248, 249 120. Odorous Principle in Milk, 249, 2n0 121. Summary of the General Constituents of Milk, . . . 250 Gases of Milk. Section Page 122. The Author's Investigation of the Gases of Milk, . . . 251-253 " Fore " Milk. 123. "Fore" Milk— Fractional Milking, 253,254 Human Milk. 124. Composition of Human Milk, 255, 256 Milk of othek Mammals : Lactescent Products of Birds and Plants. 125. Milk of the Ass, 250,257 126. „ Goat, 257, 258 127. ,, Mare, 258 128. ,, Sheep, „ . . 259 129. ,, Camel, 259 130. ,, Llama, 259 131. ,, Hippopotamus, 259, 260 132. „ Sow, 260 133. ,, Bitch, 260 134. ,, Cat, 261 135. Milk-like Secretions of Birds and Plants, .... 261 Abnormal Milks. 136. Abnormal Lacteal Secretions of Men and Animals — General Examination and Analysis of Milk, 263, 264 137. Classification of Processes for Milk Analysis, . . 264, 265 I. Microscopical and Biological Examination of Milk. 138. Microscopical Appearances of Milk — Bacteriology of Milk, . 265-267 II. Analytical Processes more particularly for the Purposes of the Food Analyst. 139. Early Processes of Milk Analysis, 267, 268 A. General Analysis of Mill:. 140. Specific Gravity— Total Solids— Extraction of Milk - Fat- Extraction of Milk-Sugar, Albuminoids, and Ash, . . 268, 269 B. Various Methods Proposed for Extracting the Milk-Fat. Solvents for Fat — Addition of Sand — Adams' Method — Esti- mation by Centrifugal Machines — Soxhlet's Process — The Werner-Schmidt Method of Fat Estimation, . . . 269-275 C. Various other Methods of Milk Analysis. 141. Drying in a Vacuum — Direct Determination of the Water — Absorption of the Water by Dehydrating Agents — Bitthausen's Copper Process— M tiller's Process — Claus- nizer and A. Mayer's Process— Summary, . . . 275-280 III. Special Details as to the more Exhaustive and Scientific Analysis of Milk. Section Page 142. Analysis of the Milk-Fat and Examination of the Ethereal Extract of Milk, 280, 281 14,S. Detection and Estimation of the Carbo-hydrates of Milk, .281, 286 144. Estimation of the Ash of Milk, 286 145. ,, Albumen, 287, 288 146. Isolation of Galactin — Estimation of the Total Nitrogen in Milk, 288 147. Isolation of the Principles precipitated by Tannin, . . 288 148. Estimation of Urea, 289 149. „ Alcohol, 289, 290 ■ 150, „ the Volatile Acids, 290 151. „ the Total Acidity of Milk 290, 291 152. Detection of Metals in Milk, 291, 292 ,, Nitrates in Milk, ...... 292 The Milk Secreted by the Unhealthy, 153. The Chemical Characters of Diseased Milk are not markedly different from Healthy Milk, 292, 293 I. human milk, 154. Some Analyses of Hmnan Milk derived from Persons in Ill- Health, 293 II. cow's MILK. 155. Milk from Cows suffering from Aphthous Fever, Mammitis, Parturient Apoplexy, Pneumonia, Engorgement of Rumen, &c. , Phthisis, and Local Affections of the Udder, . . 294-298 156. Milk of Cows suffering from Typhus, 298 157. Tlie Propagation of Disease through Milk, .... 298 The Relation of Milk to Scarlatina, 298-300 Plithisis and Tubercular Maladies, . 300-303 The Communication of Aphthous Fever by Milk, . . . 303, 304 A new Form of Febrile Disease associated with Milk, . . 304, 305 Decomposition of Milk. 158. Lactic and other Fermentations of Milk— Tyrotoxicon, . 305, SOG 159. Blue Milk, ' . . , . 306, 307 Adulteration of Milk. 160. General View of the Adulterations of Milk 307 Calculations relative to the Water in Milk, .... 308, 309 161. Method of Calculating the Removal of Cream— The Addition of Cane-Sugar to Milk, 309-311 161«. The Adulteration of Fresh Milk with Condensed Milk, . 311 162. Mineral Adulterants of Milk— Boracic Acid or Borax in Milk — Quantitative Estimation of Boric Acid— Formaldehyde in Milk— The Addition of Glycerin, Salicylic Acid, &c., 311-316 b XVlll CONTENTS. Preservation of Milk. Section Page 163. The Principles of Preserving Milk, .... ^317 164. Evaporating Processes, various Patents, .... 317, 318 165. The Preservation of Milk b}' various Additions, such as Sugar, &c., 31S, 319 166. Action of Cold on Milk, 319, 320 167. Heatmg and then Cooling Milk, as a Method of Preservation, 320 Influence of Food on the Quality and Quantity of Milk. 168. The Experiments of Weiske, Dumas, Beusch, Boussingault, Payen, Liebig, Fleischmann, and Struckmann on Feeding Animals, and the Influence of Food on the Produce of Milk, 320-324 169. Contamination of the Milk by Poisonous Colouring or Bitter Principles consumed by the Cow — The Influence of Pas- tures manured with Sewage, ...... 324, 325 170. Metals in Milk, 325, 326 The Quantity of Milk given by the Cow, the Method OF Feeding, &c. 171. The Capacity of the Udder for Milk 326 172. Relative Value of various Breeds of Cattle as Milk-Producers, 326, 327 173. The Feeding of Milking Cows, 327, 328 Cream. 174. Composition of Devonshire Cream, of Ordinary Cream, Arti- ficial Cream, 328.330 Skim -Milk. 175. Composition of Skim-Milk, 331 Condensed Milk. 176. Composition of various Skim-Milks — Condensed Milks, . 331-333 Koumiss. 177. Preparation and Composition of Koumiss, .... 333 Legal Cases Relative to Milk. 178. Defence that Milk had been deprived of Cream by Uninten- tional Skimming— The Manufacture of Condensed Milk — Novel Defence — Adulteration of Milk with Cane-Sugar and Water — Defence that the Milk was Watered by the Rain — Conviction for selling "Fore" Milk — Coaviction for selling Diseased Milk, 334-.337 Bibliography relative to Milk, ...... 337-339 Butter. 179. Constituents of Butter, 340-341 Oleo-Margarine— Butterine. 180. Manufacture and Composition of Butterine, .... 341-343 Analysis and Adulteration of Bdttee. Section Page 181. The Adulterations and General Analysis of Butter, . . 343-346 182. Certain simple Tests, 346-348 183. Cohesion Figures — Empirical Tests for Butter, . . . 348-350 184. Methods of taking the Meltiug-Point and Specific Gravity of Butter, 351, 352 Disc ]\Iethod of taking Melting Points, .... 352-354 The Titer Test, 354 Application of the Refractometer to the Testing of Butter- Fat, 354-357 The Oleo-Refractometer of MM. Amagat and Jean, . . 358 Specific Gravity, 359, 360 The Yiscometry of Butter, 360 185. Direct Titration of Butter-Fat by Koettstorfer's Method, . 361, 362 186. Decomposition of Butter-Fat into Fatty Acids and Glycerin, 362-364 Reichert-Meissl's Process, 365-367 The Process of Mr. West-Knight — Saponification by Sul- phuric Acid — Wm. Johnstone's Process — Further Analysis of the Insoluble Fatty Acids, 367-370 The Estimation of Butyric Ether, 371 Estimation of Glycerin, 371 Hubl's Iodine Method, 372-374 The Cryoscopic Method of examining Butter-Fat, . . . 374-376 Summary, .......... 376 Legal Case relative to Butter, 376, 377 Bibliography relative to Butter, 377, 378 Buttermilk. 187. Composition of Buttermilk, 378 ■ Cheese. 188. The Principles of the Manufacture of Cheese, . . . 379 189. Neufchatel Cheese — Fromage de Brie— English Cream Cheese — Camembert — Roquefort Cheese — American, Cheddar, Dunlop, Gloucester, Stilton, Gruyei-e, Gorgonzola, Skim Cheeses 379, 380 190. Parmesan Cheese, , . . . 381 191. The Ripening of Cheese, 381-383 192. Analysis of Cheese— Adulterations of Cheese, . . . 383 Arsenic in the Rind of Cheese, ...... 385 Tyrotoxicon, 385 Whey 385 Bibliography relative to Cheese, 386 Lard. 192a. Varieties of Lards, 3S6, 387 1926. Physical Characteristics of Lard, 387 192c. Chemical Characteristics of Lard, ..... 3SS-390 192d. Adulteration of Lard— Chemical Constants— Special Tests for Vegetable Oils— Detection of Beef Stearin— Larderine —Bibliography, 390-394 COKTB^fTS. PART v.— TEA, COFFEE, COCOA. I. TEA. Section Page 193. Varieties of Tea, 395 194. Structure of the Tea Leaf, 395, 396 195. Chemical Composition of Tea, 397-400 Theine or Caflfeine, Bohcic Acid, Quercitannic Acid, Quercetin, 400, 401 Composition of Tea. 195. General Analysis of Tea — Konig's, DragendorfF's, and Mulder's Analyses of Tea— Preliminary Examination of Tea, . . 401-403 Microscopical Methods of Detecting Adulterations in Tea. 197. New Process for the Examination of Leaves and Vegetable Tissues generally under the Microscope — Permanganate Process — Skeleton Ashes, ...... 403, 404 198. Chemical Method for Detection of Foreign Leaves in Tea — The Detection of Facing, 404-406 Leaves Used, or Supposed to be Used, as Adulterants of Tea. 199. Micro-structure of the Beech Leaf, Hawthorn, Camellia Sassanqua, Sloe, Chloranthus Inconspicuus, . . . 407-411 Chemical Analysis of Tea. 200. Hygroscopic Moisture of the Tea Leaf, 411, 412 201. The Estimation of Theine or Caffeine 412-414 202. Determination of Total Nitrogen, 414 203. „ Tannin, 415-418 204. The Extract of Tea 418-419 205. The Ash of Tea, 419-422 206. Determination of Gum, 422 207. General Pteview of the Adulterations of Tea, . . . 422, 423 208. Bohemian Tea, 423, 424 Bibliography relative to Tea and Theine 424 Mate. 209. Manufacture, Description, and Analysis of Mat^— Analysis of Ilex Paragiiayensis, ........ 425-427 II. coffee. 210. Description of the Coffee Berry — Microscopical Structure, . 428, 429 211. Chemical Changes during Roasting, 429, 430 212. Constituents of Coffee, 430-433 213. Analysis of Coffee, 433-435 Adulterations of Coffee and their Detection. 214. The Adulteration of Coffee with Chicory— Its Detection — Microscopical Detection of Adulterations in Coffee — Vegetable Ivory— Date Stones— Leguminous Seeds, . 435-440 CONTENTS. Xxi Section Page 215. The Adulteration of Coffee with the Seeds of Cassia, . . 441 Composition and Analysis of "Date" Coffee, 216. Estimation of Chicory in Coffee, ...... 217. The Methods of Hiepe, Prunier, Hager, and others for detect ing Adulterations in Coffee, Bibliography relative to Coffee, 442 442-445 446 447, 448 III. COCOA AND CHOCOLATE. 218. Microscopical Structure of the Seed, 448 219. Varieties of Cocoa, 450 220. Chocolate, 450 221. Average Chemical Composition of Cocoa— Cocoa Butter — Bjorklund's Test— Hager's Aniline Test, .... 451, 452 222. Composition, Effects, and^Estimation of Theobromin, . . 452-455 222a. Cocoa Red— Zipperer's Method of Determining Cocoa Red and the Products of its Decomposition, .... 455-457 2226. Determination of Crude Fibre, 457 223. The Ash — Mineral Constituents of Cocoa — Bensemann's An- alyses of Cocoa, 457-462 224. Nitrogenous Constituents of Cocoa, 463 225. Adulterations of Cocoa, , 463 226. Adulterations of Chocolate, 464 Bibliography relative to Cocoa and Chocolate, . . . 464, 465 PART VI.— ALCOHOL, SPIRITS, LIQUEURS, FERMENTED LIQUORS AND WINE. ALCOHOL. 227. Alcohol— Ethylic Alcohol, 466, 467 228. Rectified Spirit— Proof Spirit, 467 Specific Gravity Table, . 468-470 229. Tests for Alcohol 470-472 230. Separation of Alcohol from Animal Matters, . . . 472, 473 Estimation of Alcohol in Spirits and Alcoholic Liquids. 231. Various Methods for the Estimation of Alcohol — Methods used in the Examination of Alcohols in the Municipal Laboratory, Paris, 473-475 232. Estimation of Alcohol by Distillation — Tabarie's Method — Geissler's Vaporimeter — Oxidation into Acetic Acid, . 476-479 233. Methylic Alcohol, Tests for, and Estimation of — Formic Acid — Acidity of Alcohols— Aldehydes— Furfurol — Acetone — Ethers, 479-483 234. The Higher Alcohols, Bardy's Method— Bases, . . . 483-488 Brandy. 235. Composition and Adulterations of Brandy, .... 489 Rum. 236. Composition and Analysis of Rum 490, 491 Whisky. ■Section Page 237. Composition and Adulteration of Whisky, . . . .491, 492 238. EfiFects of the Higher Alcohols, 492 239. Prosecutions for Adulterated Whisky 492, 493 Gin. 240. Composition of Gin, 493 241. Oil of Calamus, 493 242. Oil of Cardamoms, 493, 494 243. Angelica Root and its Active Constituents, .... 494 244. Oil of Coriander, 494, 495 245. Oil of Juniper, 495 246. Analysis of Gin, 495, 496 Arrack. 247. Composition of Arrack, 496, 497 LIQUEURS OR CORDIALS. 248. Composition of Cordials or Liqueurs, 497 Absinthe. 249. Composition, Adulterations, and Effects of Absinthe, . . 497, 498 FERMENTATION— FERMENTED LIQUORS. 250. General Principles of Fermentation, 499, 500 251. Yeast, 500 502 252. Lactic Acid and other Ferments, ...... 502, 503 Beer. 253. Enumeration of the various kinds of Beer — Their Composi- tion, 503-505 254. The Water used by the Brewer, 505, 506 255. Malt Extract, 506-508 256. The Colouring-Matters of Malt, 508, 509 257. Beer Bitters, 509, 510 258. Hops, 510-514 259. Absynthin, 514 260. Aloin, 514, 515 261. Cnicin, 515 262. Daphnin, 515, 516 263. Gentianin 516 264. Gentiopicrin, 516, 517 265. Menyanthin, 517 266. Quassiin, 517, 518 267. The Ash of Beer, 518 268. Analysis of Beer, 518 Specific Gravity of Malt Extract, 522-525 Spirit Indication Tables, 520-528 Hop Resin and Glycerin, 531 The Nature of the Bitter used, 532 Dragendorff's Process " . . . • 532-535 CONTENTS. XXIU Section Pago 269. Adams' Process for Separating Hop Bitter — Processes to be Used for the Separation of Picrotoxin and other Matters, 536-53S 270. Special Tests for Picric Acid, 538, 539 271. Spectral Analysis of Beer, 539, 540 Salicylic Acid, 540 272. Ash of Beer— Determination of Salt in Beer, . . . 540-543 273. Adulteration of Beer with Sugar, 543 Bibliography relative to Beer, ...... 544 Wine. 274. Constituents of Wine, 545 275. Changes taking place in Wine through Age, .... 546 276. Adulterations of Wine, 546, 547 277. Analysis of Wine— Dupre's Analyses of Wines — Schmidt's Analyses of high-priced Wines — Physical Characters, Constituents Volatile below 100° C, .... 547-553 278. Volatile Acids— Estimation of free Sulphurous Acid and of Aldehyde Sulphurous Acid, 553-555 278a. Estimation of Esters in Wine, 556, 557 279. Extract or solid Residue 557-561 280. Estimation of Succinic Acid and Glycerin, .... 561, 562 281. Estimation of Tartaric Acid, and Glycerin, and Bitartrate of Potash, 562-564 Estimation of Malic Acid, 563 281a. Astringent Matters, 564 282. Estimation of Colouring-Matters in Wine, .... 564-569 283. Absorption Bands of the Colouring-Matters of Wine, . . 569 Gautier's Process for Detecting Foreign Colouring- Matters IN Wine. 284. Preliminary Preparation of the Sample, and Table A, . . 570 Table B— Systematic Process, 570-575 Special Reactions for the Detection op Certain, of the Colouring-Matters mixed with Wines. Brazil Wood, 575 Logwood, Cochineal, Fuchsine, Portugal Berries, Hollyhock, Beetroot, Elder, Privet, Whortleberries, Indigo, . . 576-578 285. Mineral Substances or Ash of Wine, 578, 579 286. Detection of Fluoborates and Fluosilicates, .... 579 Bibliography relative to Wine, 580 PART VII.— VINEGAR. 287. Constituents of Commercial Vinegar, 581 288. Adulterations, 582 289. Analysis of Vinegar, f eo Bibliography relative to Vinegar, 589 Lemon Juice and Lime Juice. Section _ Vase 290. Constituents of Lemon and Lime Juice, .... 590 291. Adulterations of Lime Juice, 590, 591 292. Analysis of Lime Juice, 591, 592 PART VIII.— CONDIMENTS: MUSTARD, PEPPER, &c. INIUSTARD. 293. Varieties of Mustard — Microscopical Structure of the Seed, . 593-595 294. Analysis of Mustard, 596 295. The Chemistry of Mustard, 596-598 296. The Fixed and Volatile Oil of Mustard, .... 598, 599 297. Adulterations of Mustard, 599-601 Pepper. 298. Varieties and General Composition of Pepper, . . . 603 Microscopical Structure of Pepper, ..... 604-606 299. Piperin — Piperidin — Piperic Acid, 606, 607 300. The Ash of Pepper— Nitrates and Nitrites in Pepper, . . 607, 608 301. General Composition of Pepper, ...... 608 302. Analysis of Pepper, 609 303. Adulterations of Pepper, 609 Olive-Stones — Poivrette, ....... 610 Legal Case relative to Pepper, 612 Bibliography, 612-614 Cayenne Pepper. 304. General Description of Cayenne, 614 305. Capsaicin and other Principles of Pepper, . . , .615,616 306. Adulterations of Cayenne Pepper, 616,617 The Sweet and Bitter Almond. 307. General Description of Almonds, 617 308. Oil of Almonds, 617, 618 309. Amygdaline— Volatile Oil or Essence of Almonds, . . 618-620 Annatto. 310. General Characters of Annatto, 621 311. Chemical Composition of Annatto, 621,622 312. Adulterations of Annatto, 622 313. Analysis of Annatto, i 623 Olive Oil. 314. General Description of Olive Oil, 623-625 315. Adulterations of Olive Oil— Refraction of Olive Oil— Tests for Arachis. Sesam6, Cotton Seed, Rape, and Poppy Seed Oils, 625-62S CONTENTS. PART IX.— WATER. Section Page 316. Introduction, 629 I. EXAMINATION BY THE SENSES. 317. Colour— Smell— Taste 629-631 II. PHYSICAL EXAMINATION. 318. Spectrum of Water, &c., 631 III. CHEMICAL METHODS. A. — Preliminary Qualitative Chemical Examination. Detection of Nitrites and Nitrates, 631-63S B. — Quantitative Analysis, 319. 1. Total Solid Residue, 633 2. Estimation of the Halogens, 634 3. Phosphates, 635. 4. Estimation of Nitrates and Nitrites— (1) Colorimetric Methods— (a.) The Brucine Method; (h.) The Diphenyl- amine Method ; (c) The Carbazol Test; {d.) Phenol and Resorcinol. (2.) Estimation as Ammonia — -The Copper- Zinc Couple — The Aluminium Process — Ulsch's JNIethod of Estimating Nitric Acid by Reduction to Ammonia. (3.) Estimation of Nitrates and Nitrites as Nitric Oxide — Crum Process. (4.) Indigo Process. (5.) Ulsch's Method of Estimating Nitric Acid by Measuring the Deficiency of Hydrogen evolved on Reduction, ..... 636-645 5. Estimation of the Dissolved Oxygen in Water, . . 646-650 6. Sulphates, ... ...... 650 7. The Forchammer, Oxygen Process, 650-654 8. Ammonia, Free and Albuminoid, ..... 654-656 9. Hardness, 65& 10. Alkalinity, 657 11. Organic Analysis of Water, . . . . . . 657 Frankland's Combustion Process, ..... 657 Blair's Method of Moist Combustion, .... 664, 665 The Author's Method of Moist Combustion, . . . 665-G67 Gravimetric Estimation of Minute Quantities of Carbon, 667-670 IV. BIOLOGICAL METHODS. 319a. Estimation of Organic Nitrogen after Kjeldahl's Method, . 670, 671 Mineral Analysis of Water, 671, 672 320. Microscopical Appearances, ....... 672 The Sedgwick-Rafter Method, 672 Mr. Dibdin's Process, 673 a. LIFELESS FORMS. Mineral Matters — Vegetable Matters — Dead Animal Matters — Human Debris, 'Manufactured Matters, . . . 676 b. LIVING FORMS. Section ^-^^ff,^ Vegetable Living Forms, 6/7-b/9 Cohn's Division of Bacteria, 679 Cultivation of Waters in Nutrient Gelatin, . . . . 682 The Sugar Test — Experiments on Animals and Human Beings, 684 Experiments on Fish, • 685 S20a. Separation and Identification of Pathogenic Organisms from Water. 1. Typhoid Bacillus. 2. Cholera Bacterium, . 686-689 321. Interpretation of Results, 689-691 Appendix to Water Analysis. 323, Standard Solutions and Reagents, &c., alphabetically arranged, 692-695 Tables for Use in Water Analysis, 696, 697 Appendix. The Sale of Food and Drugs Act, 1S75, 699-707 Sale of Food and Drugs Act Amendment Act, 1879, . . 708, 709 The Margarine Act, 1887, 709-711 New York Adulteration Act, 1881, . . - . . 711-713 Index, 715 LIST OF TABLES. Table . Page I. Action of volatile solvents on colouring-matters in acid solution, to face 117 II. Action of volatile solvents on colouring-matters in alkaline solution, toface 117 III. Action of acids on colouring-matters, . . ,, 117 IV. Solubility of sugar in alcohol of different strengths, . 12S V. Composition of raw beet-root sugars of good quality, . 133 VI. Analyses of concrete sugars, ....-• 134 VII. Table showing the loss by volatilisation (according to Landolt's experiments) of sugar ash, .... 135 VIII. Sugar in parts per 1000 corresponding to cc's. of the ammoniated copper test, 145 IX. Some adulterated samples of honey analysed by Dr. Sieben, 156 IXrt. Some analyses by Dr. Wallace of molasses, treacle, &c., 158 X. Composition of various fruits, . . . • .159 Xa. Giving the results of estimations of starch by different methods, 169 XI. Karmarsch's and Wiesner's measurements of starches, . 178 XIa. Weinwurm's analyses of flour, ..... 184 XII. Composition of infauts' farinaceous foods, . . . 209 XIII. Showing the percentage of potato starch and dry sub- stance corresponding to various specific gravities, . 221 XIV. Table giving specific gravity of ether holding fat in solution for use with Soxhlet's apparatus, . . . 276 XV. Weights of copper corresponding to different quantities of milk-sugar — analyses of commercial milks, . . 284 XVI. Showing the influence of food on the milk secretion of the cow, ......••• 3-3 XVII. Showing the influence of breed on the milk-producing powers, ... . ■ . . . . • 3-0 XVIII. Average yield of milk, 326 XXVlll LIST OP TABLES. Table Page XIX. Composition of condensed milks, 332 XXa. Table giving refractive angle of butter fat at 38°, melting point, specific gravity, &c., 357 XX6. Composition of cheese, ....... 380 XXI. Changes in Roquefort cheese through age, . . . 3S3 XXIa. Chemical and physical constants of lard, and some oils used for the adulteration of lard, .... 391 XXII. Solubility of theine, 399 XXIII. Ash of various species of tea, 4S0 XXIV. Composition of various species of tea, .... 421 XXV. Changes in coffee during roasting, 431 XXVI. Imitation coffee berries, 436 XXVII. Theoretical quantity of soluble ash in various quantities of chicory and coffee, ....... 444 XXVIII. Specific gravity of infusions of coflfee and chicory, . . 44.5 XXIX. Composition of roasted cocoa beans after removal of the husk, 458 XXX. Konig's analyses of cocoa beans, ..... 459 XXXa. Analyses of cocoa made in the U.S. Agricultural Labora- tory, 460 XXX6. Bensemann's analyses of cocoa, cocoa husks, and choco- late, 461 XXXc. Numerical relations between tiie various constituents in l^receding table, . . ...... 462 XXXI. Nitrogenous constituents of cocoa, .... 463 XXXII. Properties of the principal alcohols, .... 466 XXXIII. Estimation of alcohol— specific gravity tables, . . 468 XXXIV. The alcoholic content of boiling spirits as shown by the temperature of their vapour, 475 XXXV. Boiling points of dilute alcohol, 475 XXXVI. Special examination of certain beers 504 XXXVII. Composition of barley and malt, 506 XXXVIII. Constituents of the ash of hop-cones, .... 513 XXXIX. Ash of beers, 519 XL. Specific gravity and strength of malt extract, . . 522 XLI. Spirit indication table, 526 XLII. „ „ ,, 527 XLIII. Table for ascertaining the strength of acetic acid in beer analysis, 528 XLIV. Composition of wine according to Dupr(5, . . . 548 XLIVa. Schmidt's analyses of certain fine and exceptionally valuable old wines, 550 XLV. Hager's table of wine extract, 559 LIST OF TABLES. XLVa. XLVI. XL VII. XL VIII. XLIX. L. LL LII. LIII. Lllla. JAIIb. LIV. LV. LVI. Page Process for the detection of the colouring-matters in ■wine, to face 570 The composition of various kinds of vinegar, . . . 582 Composition of mustard, 595 Analyses of ash of mustard seed, 596 General composition of commercial peppers, . . . 60S Chemical and physical constants of olive oil and some oils used for the adulteration of olive oil, . . . 624 Mixed fatty acid, 625 Analysis of mixtures of olive and arachis oils, . . 627 Estimation of chlorine as common salt, . . . 6.S5 Value of indigo in nitrogen for different strengths of nitre solution, ........ 642 Analysis of the Grand Junction water for the year 1881, 647 Microscopical examination of water from Wannacomet Pond, Nantucket, 674 Hardness in grain measures, ...... 695 ,, in parts per 100,000, 696 Reduction of cubic centimetres of nitrogen to grammes, 697 LIST OF ILLUSTRATIONS. Fig. Page 1. Soxhlet's apparatus for extraction of fat, .... 67 2. Clausnizer's modification of Soxhlet's apparatus, ... 68 3. Simple form of fat extractor, ....... 68 4. Wynter Blyth's tube for the treatment of liquids by volatile solvents, .......... 69 5. Wynter Blyth's ether-recovery apparatus, .... 70 6. The mercury pump, ......... 70 7. The Sorby-Browning micro-spectroscope, 74 8. Diagram for the reduction of spectroscopic lines and bands to wave lengths, ......... 77 9. " Hema-spectroscope comparateur " of Maurice de Thierry, . 78 10. A simple apparatus for obtaining spark spectra, ... 80 11. The polarising colorimeter, ....... 81 12. Colorimeter with the Lumner-Brodhuns prism, ... 83 13. Vierordt's divided slit, ........ 84 14. Kriiss's universal spectral apparatus, ..... 87 15. Stein's method of obtaining micro-photographs, ... 89 16. Diagram representing various spectra, ..... 94 17. ,, ,, absorption spectrum of logwood and alum, 9S 18. ,, ,, ,, ,, of aniline greens, . 106 19. Flask for determining carbon dioxide, ..... 120 20. The saccharimeter of Mitscherlich, ...... 146 21. Microscopical structures of the damson, ..... 160 22. ,, „ of the plum, 160 23. ,, ,, of the rind of orange, . . . . 161 24. Pulp cells of strawberry, . . . . . . . . 162 25. Structures found in the currant, ...... 162 26. Microscopical structure of wheat, ISO 27. ,, ,, of corn-cockle, 187 28. „ ,, of the oat, 210 29. „ „ of barley, 212 30. „ „ of rice, 215 LIST OF ILLUSTRATIONS. XXxi Fig. Page 31. Microscopical structure of maize, 216 .32. Diagram of the potato fungus, 219 33. The lactocrite, 272 34. Soxhlet's apparatus for the rapid analysis of milk, . , . 274 35. Reinhardt's apparatus for taking melting points, . . . 351 36. Amagat and Jean's oleo-refractometer, 358 37. Flask method of washing the fatty acids, 363 3S. Muter's oleia tube,* 369 39. Cryoscopic apparatus, 375 40. Figure of the tea plant, 397 41. Idioblasts in the tea leaf, 397 42. Epidermis of tea leaf, showing micro-structure, . . . 398 43. Skeleton ash of tea leaf, 404 44. ,, ,, of sloe leaf, 404 45. ,, ,, of lime leaf, 404 46. ,, ,, of tobacco leaf, ....... 404 47. Epidermis of beech leaf, 407 48. „ of hawthorn leaf, 408 49. ,, of camellia sassanqua, ...... 409 50. Section of sloe leaf, 409 51. Epidermis of under surface of leaf of the chloranthus incon- spicuus, 410 52. Endosperm of coffee, 428 53. Membrane from coffee berry, 428 54. Coffee tissue, 429 55. Regnault's volumemometer, ....... 434 56. A section of the carob-tree fruit, 437 57. Structure of the vegetable ivory nut, 438 58. ,, of the date-stone, 439 59. Section of the seed of parkia, 440 60. ,, of seed of cassia occidentalis, 441 61. Tissues of cocoa, 449 62. The distillation of spirits, 476 63. Geissler's vaporimeter, ........ 477 64. Sell's tube for estimation of amyl alcohol, .... 484 65. Wynter Blyth's tube for sediments, 502 66. A representation of various ferments, 503 66. Diagrammatic representation of various spectra, . . . 569 67. Section through the coats of sinapis alba, .... 593 68. Tissues of mustard, 594 69. A section of black pepper, 605 70. Ground pepper, ......... 606 71. Tissues of the olive stone, 610 XXxii LIST OF ILLUSTRATIONS. Fig. Page 72. Apparatus for use in Ulsch's method of estimating nitric acid, 643 73. The mercury pump, 659 74. Apparatus for gas analysis, 660 75. Gas pipette .... 661 76. Apparatus for moist combustion, 666 77. Tube for sediments, 672 78. Closterium 678 79. Diatom vulgare, ......... 678 80. Various forms of infusoria, ....... 681 81. Drop-bottle for cultivation, 682 PLATES. Frontispiece — Microscopical characters of Lard and Beef crystals. Plate of Soleil's Saccharimeter, . . . to face page 147 Illustrations of Organisms supposed to be the cause of Scarlet Fever to face page 299 Photo-illustration — Drops of Tallow and Spermaceti, to face page 347 P^RT I. HISTORY OF ADULTERATION. FOODS: THEIR COMPOSITION" AND ANALYSIS, PART I.- HISTORY OF ADULTERATION. I.— EARLY NOTICES OF ADULTEKATION, ESPECIALTA" IN ENGLAND. § 1. Before adulteration commences, commerce must develop. In primitive states of society,* there may be knavish tricks, ignorant bartering, substitutions of bad for good, falseness and meanness of all kinds, but no systematic sophistication is possible. Again, in the semi-pastoral state (as it existed in some parts of Scotland a century ago), in which the food of a family is raised from the soil on which they dwell, and clothing produced from their own sheep and spun into textile garments at their own. fireside, commercial frauds are imknown or undeveloped. There are several notices of ancient sophistications practised by the Greek and Roman traders ; but it is from the Middle Ages that the most copious and interesting materials for a history of adulteration are obtained — a page of histoiy but little explored, yet abounding with curious facts more or less illustrative of the manners of the times. The mixing, or, rather, alloying of gold or silver with the baser metals, may be justly considered of the nature of adultera- tion, and has prevailed contemporaneously with the ai-t of coinage. The well-worn tale of the detection of the base metal in the crown of Hiero by Archimedes, some two and a half centui-ies before Christ, may be accepted as probably the earliest scientitic detec- tion of adulteration. The process iised by the philosopher of Syracuse when discharging the duties of a jmblic analyst, and now called specific gravity, is quantitative as well as qualitative, and, * "High things begia low. Astronomy began as astrology, and when trade began, there must have been even more trickerj^ about it than there is now. Conceive a world made up of nomadic tribes engaged in perpetual warfare. It is a commerce of killing. If a tribe desire the richer soil or larger possessions of another, the method is to exterminate that other ; but at last there rises a tribe too weak or too peaceful to exterminate, and it prefers to barter, it challenges its neighbours to a contest of arts. They try to get the advantage of each other in bargains, they haggle and cheat ; it is not heroic at all, but it is the beginning of commerce and j)eace." — "Demou- ology," by Maiu'ice Conway, M. A. 4 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 1. thougli purely physical, is used daily by all engaged in practical chemistry. Vitruvius* in his work on architecture describes the adultera- tion of minium with lime. He also gives a simple process for its detection : heat to redness on a sheet of iron ; if pure, it will blacken, but on cooling return to its former hue. Dioscorides alludes to the adulteration of opium with gum and with the milky juice of glaucium and lactuca. The test for dis- tinguishing the pure fi-om the false was primitive : the opium was to be burnt ; if pure, the flame was clear and brilliant, but the adulterated burned with difficulty. The quality of the opium was also to be judged by its behaviour when exposed to the rays of the sun ; when opium of good quality liquefies, it looks as if it had just come from the plant. + He also specifies the adultera- tions of several drugs ; as, for example, the mixing of styrax resin with styrax sawdust. Pliny, among other matters, alludes to the frauds practised by bakers ; for they added to the bread a white earth, soft to the touch and sweet to the taste, which was obtained from a hill called " Leucogee " situated between Puzzoles and Naples. It has been suggested that the white earth was carbonate of magnesia ; this is doubtful. % He also speaks of the adulteration of aerugo (under which name was confounded both the acetate and sulphate of copper) with shoemaker's black atramenta sutorium, and gives a true chemical method for its discovery. Paper is to be soaked in the juice of galls ; if the aerugo is pure, it will not turn the paper black. Another method was to put the substance on a sheet of red hot iron ; if sulphate of iron had been added, it became covered with spots. § The adulteration of wine in Athens necessitated the appoint- ment of a special inspector, whose duty it was to detect and stop these practices. Gi-eek history has handed down the name of one Canthare, who excelled in ingenious mixtures, and knew how to impart the flavours of age and maturity to new wines. His ingenuity was such, that it was commemorated in the proverb : " Artificial as Canthare." In Piome, also, wine was much tampered with ; even the rich, according to Pliny, could not obtain the natural wines of Falei'no, for they were adulterated in the cellai-s ; and certain wines from Gaul had an artificial colour given to them by means of aloes and other drugs. || * Vitruvius, i. ix. c. 13. t Dioscorides, iv. 65; Pliny, xx. 76. ^ riiny, xvfii. 29. § PHny, xxxiv. 1| Op. cit. § 2, 3.] EARLY ADULTERATION IN ENGLAND. 5 § 2. In our own country, and in Europe generally, from the eleventh century onwards, the bakers, the brewers, the " pepperers," and the vintners, were most frequently accused of corrupt pi-actices. We must not, however, judge too harshly of the tradespeople of that epoch, for morality was generally low, and adulteration an innocent pastime when compared with the frequency and magni- tude of midday highroad robbery and midnight violence. In the latter part of the twelfth century, that which would now be called crime became the favoui'ite amusement of the principal citizens, "who would sally forth by night, in bands of a hundred or more, for an attack upon the houses of their neighbours. They killed without mercy every man who came in their way, and vied with each other in brutality. . . , False weights, false measures, false pretences of all kinds were the instruments of commerce most generally in use. No buyer would trust the word of a seller, and there was hardly any class in which a man might not with reason suspect that his neigh- bour intended to rob or even to murder him."* ASSIZES OF BREAD— BAKERS. § 3. The sale of bread was regulated in England as early as the fourth year of the reign of John, by what was called the " Assize of Bread," the original object of which was to regulate the price of bread by limiting the profit of the baker on each, quarter of wheat, so that the price of the loaf should bear a certain proportion to the price of the quarter of wheat. The assize of John's reign continued in force until 128G, when it was repealed by " The Statute of Assize." There were various modifications of these assizes, and they were finally abolished in 1815. The "Assize of Bread" in its influence was probably the exact reverse of what was intended. On the one hand, the development of trade was restricted inju- diciously, and, on the other, the bakers often suffered unjustly, and, therefore, had a direct inducement to recover their losses by nefarious practices. Although, at the institution of the assize, adulteration with foreign substances was not the main object of the regulations, yet, as time went on, and the sins of the bakers, both male and female, f accumulated, clauses with regard to the * " A History of crime." By Luke Owen Pike. t The bakers, as well as tlie brewers, were of both sexes. 6 FOODS : THEIR COUrOSITIOX AND ANALYSIS. [§ 3. adulteration of bread were inserted, and the later ones may be considered collectively as tlie ancient English " Sale of Food Act." The assize of 1582* contained the following : — " If there be any that by false meanes useth to sell meale : for the hrst time he shall be grievously punished, the second tynie he shall lose his meale : the III tyme he shall foreswere the towne, and so like- ■wyse the bakers that oticnde. Also, bouchers that sell mesell porke or uiozen flesche : for the first time they shall be grievously amei'ced, for the second tyme so oil'endinge they sliall have the judgement of the pillory, for the third tyme they shall be oomytted to pi-yson until I'ansomed, and the fourth tyme they shall forswere the towne, and thus ought other transgressors to be punished, as cooks, forestallers, regrators of the markets when the cookes serve, roste, bake or any otherwyse dresse, fysclie or flesche unwholesome for man's body." The assize of 163-i had some stringent regulations with regard to musty meal : — " If there be any manner of person or i)ersons, which sliall, by any false Avayes or meanes, sell any meale unto the kinge's subjects, either by mixing it deceitfully or [sell any] musty or corrupted meal, which may be to the hurte and infec- tion of man's body, or use any false weight, or any deceitful wayes or meanes, and so deceive the subject, for the lirst ofience he shall be grievously punished, the second he shall lose his meale, for the third offence he shall suffer the judgement of the pillory, and the fourth time he shall forswere the towne wherein he dwelleth." These extracts give some idea of the punishments inflicted on dishonest bakers during the Middle Ages in England. First oftences were often visited by corporal chastisement and exposure in the pillory (generally with a rope and a loaf round the neck) ; foui'tb, and even third, convictions were considered so heinous that it was thought better to cast the man forth from the city to earn his livelihood elsewhere. In tiie curious paper entitled "A Quip for an Upstart Courtier,""}* there is a powerful and quaint exjiostulation with the diflerent traders : — " And for you goodman baker, you that love to be seen in the open market-place upon the pillory, the v/orld cries out on your wiliness : you crave but one deere yeare to make your daughter a gentlewoman. You buy your corne at the best hand, * The title runs : — "Here beginneth the boke named the assj-se of bread, what it ought to weye, after the pryce of a quarter of wheat, also the assize of ale, with all maimer of wood and cole, lath, bolside, and tymber, and the weight of butter and cheese. Imprynted, by Thomas \V3'att, 1582." t The "Quip for au Upstart Courtier " was written in 1592. The original is in black letter. § O.J EARLY ADULTERATION IX EXGLAKD. ,7 and yet will not be content to make your bread weight by many oiinces. You put in yeaste and salt to make it lieavie : and yet all your policie cannot make it. The poore crie out. the riche find fault, and the lord maior and the .sheriffs, like honourable and worshipful maiestrates, every daie walk abroad and weigh your bread, and yet all will not serve to make you honest men. But were extremities used and the statutes put in the highest degree in practice, you would have as few eares on your heade as the colly er."' The manner of adulteration seems to have varied, f Some- times the bread wa« made altogether of " putrid and rotten materials," sometimes it was good outside and bad within, and as for the addition of alum or mineral matters, such was only detected "when in considerable quantity and coarsely done. The more artfid mixtures required for their detection the application of a chemical science not then possessed. The following may serve as examples of a few of the earlier instances : — One '"Alan de Lyndseye, baker, was sentenced to the pillory because he had been convicted of baking pain deniayn that was found to be of bad dough within and good dough without, and because such falsity redounds much to the deception of the people who buy sucli bread." The same baker seems a few days afterwards to have been again in trouble, for "Alan de Lyndseye, bakei-, and Thos. de Patimere, baker, were taken and brought before the Mayor and Aldermen, and sentenced to the pillory for selling bread made of false, putrid, and rotten materials, through which who bought bread ■were deceived, and might be killed.'"' % A similar fraud is recorded at perhaps an earlier date (a.d. 1311), for "the bread taken from "William de Somersete, baker, ■ on the Thursday next before the Feast of St. Lawrence (Aug. 10) in the oth year of the reign of King Edward II. was examined and adjudged upon. . . . Because it was found that such bread was putrid and altogether rotten, and made of putrid ■wheat, so that persons eating that bread would be ])oisoned and choked, the Sheriff was ordered to take him and have him here on Friday next after the Feast of St. Lawrence, then to receive judgment for the same."§ * The sanitary state of the bakehouses in the fifteenth and sixteenth centuries was, as a rule, bad. According to Mr. Pike, they appear to have been the favourite receptacle for dead bodies after a murder had been com- mitted. " A History of Crime," by Luke Owen Pike, M.A. Vol. i., p. 256. t In the reign of Edward I. it was enacted at one of the " Hallmots," •that no bread should be coated with bran, or so as to be found worse when broken than it was on the outside. % " Memorials of London," by II. T. Riley, pp. 120, 121. § Op. ciL 8 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 4. Mr. Pike, in his " History of Crime,"* speaks of loaves being adulterated in the Middle Ages by lumps of iron, probably refer- ring to the following case. " On Wednesday next after the Feast of St. Matthew the Apostle (11th of Sept., a.d. 1387), in the 11th year, Richard Porter, servant of John Gibbe, baker, of Stratforde, was brought here before Nicholas Extone, j\Jayor of the said city, John Hafile, and other Aldermen, and questioned for that when the same Mayor on that day went into chepe to make assay there of bread, according to the custom of the city, he, the said Kobert, knowing that the bread of his master in a certain cart there was not of full weight, took a penny loaf, and in it falsely and fraudulently inserted a piece of iron weighing about Gs. 8d. (4 oz.), with intent to make the said loaf weigh more. . . . He was sentenced to the pillory Avith the loaf and iron round his neck, and the cause of the punishment was proclaimed by the Sheriffs."! But the placing of a mass of metal in a loaf under the circumstances recorded is somewhat different from adulteration, for the man slipped in the iron to avoid a conviction for false weights. BREWERS AND VINTNERS. § 4. Beer. — The fraudulent practices of the early brewers are thus detailed in the Black-letter Tract before mentioned : " And you, maister brewer, that growe to be worth forty thousand pounds by selling of soden water, what subtilty have you in making your beere to spare the malt, and put in the more of the hoppe, to make your drinke, be barley never so cheape, not a whit the stronger, and yet never sell a whit the more measure for money. You can when you have taken all the harte of the malt away, then clape on stoi'e of water, t'is cheape enough! and mashe out a turning of small beere, like rennish wine : in your conscience how many barrels draw you out of a quarter of malt ? Fie ! fie ! I conceal your falsehood, least I should be too broad in setting down your faults." Not only the brewer but the retailer of the beer, was also condemned. " Last to you, Tom Tapster, that take your small Cannes of beere, if you see your guests begin to be drunke, halfe smal and halfe stronge ; you cannot be content to pinch with your small pottes and your ostrie faggots, but have your drugges and draw * Vol. I., p. 237. t Riley's " Memorials of London," p. 493. § 5.] EARLY ADULTERATION IN ENGLAND. 9 men on to villany and to bring customers to your house, where you sell a joint of meat for xii. pence that cost you scarce six, and if any cliance to go on the shore, you shore him when he is asleep and set up a pot a day more than he hath, to find you drinking pots with your companions. To be short, tliou art a knave ! " As early as the reign of Edward the Confessor, we find it recorded in Domesday Book that in the city of Chester a knavish brewer, '■'■malani cerevisiam /aciens, in cathedra jionehatur stercoris" — in other words, the offender was taken round the town in the cart in which the refuse of the ])lace had been collected, and to this degradation was often added corporal chastisement. In many towns in the sixteenth century, we find "ale-tasters,"' whose duty it was to insi)ect tlie beer. In 1529, for example, the Mayor of Guildford ordered that the brewers make a good useful ale, and that they sell none until it be tasted by the "ale-taster." These oiBcials had to take the follow- ing oath : — "You are chosen ale-tasters of this town. You shall well and truly serve his Majesty and this town in the same office. You shall at all times tiy, taste, and assize the beer and ale to be put to sale in this liberty, whether the same be whole- some for man's body, and present those that offend, or refuse to suffer you to assay it. You shall give your attendance at all courts, and present from time to time the offenders, and all things else belonging to your office you shall do and execute. So help you God." The ale was not only tasted, but some of it ■was spilt on a wooden seat, and on the wet place the taster sat, attired in leathern breeches, then common enough. If sugar had been added to the beer, the taster became so adherent, that rising was difficult; but if sugar had not been added, it was then considered that the di-ied extract had no adhesive property. A less coarse, but not dissimilar, method was also applied by the earlier Inspectors to test the pui-ity of milk. § 5. Wine. — The frauds of the vintners or winesellers attracted some share of public attention in the sixteenth and seventeenth centuries, as shown by municipal records, fugitive tracts and broadsides. In August, 1553, a certain Paul Barnardo brought into the port of London some wine, and there is extant an order in council directing the Lord Mayor to find five or six vintners to rack and di-aw off the said pipes of w-ine into another vessel,. and to certify what drugs or ingredients they found in the said wine or cask to sophisticate the same.* At a later date the records * Rememhrancla, vii. G2. 10 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 6. of tlie Common Council contain a certificate from tlie Lord Mayor to the lords of the council, stating that the wines of a certain ^'Peter Van Payne" had been drawn off in his presence, and that in eight of the pipes had been found bundles of weeds, in four others some quantities of sulphur, in another a piece of match, and in all of them a kind of gravel mixture sticking to the casks; that they were conceived to be unwholesome and of a natvire similar to othei\s formerly condemned and destroyed.* In "The Search after Claret," by Richard Ames, a thin quarto, the last leaf is occupied by the following advertisement : " If any vintner, wine-cooper, etc., between Whitechajiel and "Westminster Abbey, have some tuns or hogsheads of old rich unadulterated claret, and will sell it as the law directs for sixpence a quart, this is to give notice, that he shall have more customers than half his profession, and his house be as full from morning to night as a conventicle or Westminster Hall the first day of term. "t Later, tlie vintners became more scientific in their operations. Addison (in the Tatler, No. 131, 1710) alluded to a certain fra- ternity of chemical operators who wrought underground in holes, caverns, and dark retirements to conceal their mysteries from the eyes and observations of mankind. " These subtle philosophers are daily employed in the transmutation of liquors, and by the power of magical drugs and incantations raise under the streets of London the choicest products of the hills and valleys of France; they squeeze Bordeaux out of the sloe, and draw Cham- pagne from an apple." SPICES— DRUGS. § 6. The London pepperers, or spicers, formed a separate guild, and were under special ordinances. The ordinance, A.D. 1316, in Norman-French, has the following regulations: — " No one of the trade or other person in his name for him, shall mix any manner of wares, that is to say, shall put old things with new, or new things with old, by reason whereof the good thing may be impaired by tlie old, nor yet things of one price, or of one sort, with other things of another sort; also, that no man shall dub any manner of wares, that is to say, by putting in a thing that was in another bale, and then dressing the bale up * Rememhrancia, \\n., 12th July, 16.35. t " The Search after Claret, or a Visitation of the Viutners." A poem ia Two Cantos. 2nd Ed., Loudon, 1697. 4to. § 6.] EARLY ADULTERATICX IN EXGLAND. 11 again in another manner than the former in which it was first bought, so as to make the ends of the bale contain better things than the remainder within tlie bale, by reason w^hereof the buyer may be deceived, and so lose his goods. Also, that no man shall moisten any manner of merchandise, such as saffron, alum, ginger, cloves, and such manner of things as may admit of being moistened ; that is to say, by steeping the ginger, or turning the saffron out of the sack and then anointing it, or bathing it in water ; by reason wliereof any manner of weight may, or any deterioration arise to the merchandise."* In England the trades of the druggist and the grocer were com- bined. Drugs and groceries were sold in the same shop, and they were under the same regulations until 1617, when the apothe- caries separated themselves from the grocers. Very soon after they had become a distinct body, they began to complain of the frauds and artifices of the grocers, from whom they continued to be supplied with many drugs; and, therefore, established a dis- pensary for the purpose of compounding the more important jDreparations themselves. In 1 540 the physicians were empowered to search, view, and see the apothecary- wares and stuffs, and to destroy such as they found unfit for iise. In 1553 very exten- sive powers were conferred on the College of Physicians for this purpose. " The four censors, or any three of them, shall have authority to examine, survey or govern, correct and punish all and singular physicians and practisers in the faculty of physic, apothecaries, druggists, distillers, and sellers of watei's and oils, and preparers of chemical medicines, according to the nature of his or their offences." The great power of the censors was on more than one occasion abused. In 172-1, for example, they burnt the drugs of one " Goodwin" the drugs not having been examined, and the history of the whole affair showing that the act was merely a gratification of private spite. Goodwin peti- tioned Parliament, and ultimately, it is said, obtained £600 compensation. The College of Physicians compiled the first Pharmacopoeia, and pviblished it in 1613. Subsequent editions bear the dates of 1621, 1632, 1650, &c. As may be expected, the early editions contain lists of very absurd and superstitious remedies, and have no pretensions to a scientific character. f * Riley's " Memorials of London," p. 120. + See " Historical Sketch of the History of Pharmacy." By Jacob Bell. Lond., 1861. 12 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 7, 8» II.— ADULTERATION IN FKANCK § 7, In France, from very early times, the general supervision of provisions, as to pxirity and quality, and the inspection of weights and measures, were under the ^'■police des commissaires,'"*' and various special statutes were enacted from time to time. Thus, an ancient statute (1292) of the Paris brewers forbade the adulteration of beer ; " whoever put into beer baye, pimento or '■poix resine ' was to be fined 20 francs, and his brassins were to be confiscated, for such things are neither good nor loyal to put in beer, for they are bad for the head and for the body, for the healthy and the sick." A later statute, dated March 16, 1G30, among various sanitary provisions, forbade the use of buckwheat, *' yvtoye or other bad matters under a penalty of 40 Parisian pounds." Judges were also to examine the materials before use, in order to see that there was nothing in them impure, heated, mouldy or spoiled. If such were found, the materials were to be cast into the rivei-.f § 8. Flour and Bread. — There were various special regulations as to flour and bread ; by an Ordonnance of the Provost of Paris, October 11, 1382, the miller was to grind the corn without mixing it, to increase his fee, with bran, pease, beans, or anything else save that which had been given him to grind.;}: Later, by a decree, dated July 13, 1420, the bakers were forbidden to be millers, it being thought that if they ground the wheat as well as made it into bread, there would be facilities for fraudulent deal- ing. The punishment of bakers for false bread — whether the falseness were admixture of foreign substances, the use of damaged flour, or simply light Aveight— was very similar to that of English bakers, except that it partook more of the character of a religious penance. Thus, in 1525, a baker convicted of " false * "La police des commissaires . . . il est de lews soins de /aire punir le debit des vivres corrumpus, alterez, falsifiez, lesfauxpoids et lesfaux niesures." Traile de la Police de la Mare. Tom. i. liv. i. titre xi., chap. vi. t " La police des commissaires visitoient les Marchez, et il estoit de lews soiiis d'y procurer L'ahondance des vivres et des aidres provisions neressaires a la subsistance des citoyens, Us emperhoient qu'il ne s'y commist aucunefravde, soit en la qualite ou au prix, soit au poids ou en la mesure, ils estoient princi- paleinent charges de se donner ious les soins a regard des grains, du pain, de la viande, et da vin." Loc. cit. X " Que riul meusnier ne soit si os6 ne si hardy sur quanque il se pent mejaire envers le ray, en corps, et en biens, de mesler, mettre ou fair mettre en aucune maniere es farines des hlez qu'ds moudront aucune mixtion on meslee, pour rendre plus grande moiUure, cumine de bran, d'orge, de pois, de/eves, ou autres choses qnelconques, qui ne soit du bU qui leur sera baillL" Traiti de la Police, t. ij. liv. v. titre ix. § 9.] EARLY ADULTERATION IN FRANCE. 13 bread" was condemned by the court to be taken from the Cliate- kt prison to tlie cross before the " Eglise des Carmes," and tlience to the gate of Notre Dame and to other public places in Paris, in his shirt, having the head and feet bare, with small loaves hung from his neck, and holding a large wax candle lighted, and in each of the places enumerated he was to make '■'■amende honorable," and ask mercy and pardon of God, the king, and of justice for his fault.* False weights were also often punished by corporal punishment. In 1491 the case of three bakers is recorded, who, having been convicted of selling loaves "too small," were stripped and beaten with rods through the streets of Paris, and were admonished for the future to sell the three kinds of bread ordered by the law, of the weight and quality they ought to be. t In still later times, we find the practice of the courts remarkably severe. In 1699, a baker named Pasquier, was convicted of converting into bread bad and unwholesome flour. Sacks tilled with good tiour and others filled with bad, had been found on his premises, and it was affirmed that he had mixed the two together. He was fined 500 livres, his oven de- molished, and his shop closed for six months with a placard upon it stating the crime and the punishment. % § 9. Wine. — A curious decree of the Provost of Paris, in 1371, compelled the tavern-keepers to permit any one who purchased wine, whether to be drunk on the premises or taken away, him- self to see the wine drawn from the cask. The penalty was, for neglect or disobedience to this law, four Pai'isian pounds, one- fourth of which went to any informei-. An Ordonnance of January 30, 1330, forbade the mixing of two wines together ; no wine-seller was to give a false name to a wine, or to give a wrong description of its age ; the penalty was confiscation of the wine and a fine. Similar edicts were promul- gated in 1415, 1635, and 1672. Still the evil did not diminish, and in 1708 two hundred inspectors of wine and drinks were appointed in Paris. The "Baillie" of Bergheim, in 1718, had condemned to a month's imprisonment one Andre, who had falsified his wine with some poisonous plant (probably belladonna), and his wife, who had sold the wine, to a month's imprisonment, and a fine of 130 livres. This Avine caused the death of one person, and the illness of several who had partaken of it. The sentence having been annulled on the appeal of the condemned to the Superior Council of Alsace, Andre and his wife were ultimately ordered to be led * Traile de la Police, tome ii. livre v. titre xii. t Op. cit., tome ii. livre v. titre xii. X Op. cit., t. i. livre iv. titre iv., p. 570. 14 foods: their composition and analysis. [§ 10, 11. by two sergeants for one da^y through the streets of Bergheim,. carrying placards both before and behind, with '• frelateurs cle vin" printed thereon. They had also to pay 30 livres fine,. ^' pour faire prier Dieu 2^our le repos de I'd.ne du defunt," and tho fine of 130 livres, pronounced by the first judge. The council promulgated a very severe decree directed against such practices. It was also forbidden to adulterate wine with litharge, Indian wood, isinglass, "raisin de hois," or other drugs, or mixtures capable of injuring the health of those who drank the wine, under a penalty of 500 livres and corporal punishment. Even the pos- session of matters likely to be used for adultei-ation was an offence. So late as 1710, one Denys Porch er and his wife were convicted of conveying barrels of " vin de raisin de bois " into Paris. They were fined 30 livres, the four barrels of wine were spilt on the pavement, and the sentence placarded in Paris and various places around. § 10. JJutter. — An Ordonnance'^ of the Provost of Paris, dated November 25, 1396, forbade the colouring of butter with "soucy flowers," other flowers, herbs, or drugs. Old batter, likewise^ was not to be mixed with new, but the sale was to be separate, xmder penalty of confiscation and fine. The ancient laws of the merchant butter-sellers and fruiterers, confirmed in 1412, reiterated the above, and also forbade the sale of butter in the same shop in which fish was sold. The retail or sale of butter by spicers, chandlers, apothecaries, and generally by all carrying on ofi"ensive trades, was made illegal. A subsequent enactment in 1519 confirmed this law. § 11. Drugs. — The Drug-sellers were also under regulation, and without doubt their practices, witli regard to sophistication, were q,uite on a par with those of other trades. Gargantua in Paris is made to visit the shops of druggists, herbalists, and apothe- caries, where he " diligently considered the fruits, roots, leaves, gums, seeds, the grease and ointment of some foreign parts, as. also how they did adulterate them — i.e., all the said drugs." f In the Middle Ages the French apothecaries were at fii'st confounded and amalgamated, as in England, with the merchant spicers ; but in 1777 the two trades were separated, and they formed a definite body. In the fifteenth century the shops were little more than open booths, as may be seen from a miniature in "Ze Regime des Princes," a manuscript of the fifteenth century,, preserved in the Arsenal Library, Paris. Philip YI., as early as 133G, issued a regulation by which no * Trnite de la Police, tome i. livre iv. titre ix. t " R-abelais," cxxiv., p. 53. § 12, 13.] EARLY ADULTERATION IN GERMANY. 15 one could be au apothecary unless he was a born or naturalised. Frenchman, and a good Catholic. According to the law, neither the spicers nor the apothecaries were permitted to employ in the preparation of their medicines, drugs, confections, conserves, oils, or syrups, any sophisticated or exhausted or corrupted drugs,, under penalty of a fine of fifty livres, and the seizure and burning of the merchandise thus adulterated in front of the dwelling in which it had been found. Charles VIII. released the apothecaries from some of the strict regulations of earlier times, and both he and his successors were the authors of many edicts relative to the apothectiries and spicers ; besides wiiicli these trades were regulated by local enactments in different towns. § 12. Conseils de Saluhritc. — In 1802 the "Conseil de Scdidyrite^ was established in Paris. It originally consisted of only four membei's, and took cognisance of adulteration, epizootics, un- healthy trades, and a little later of the administration of prisons- and public charities. Afterwards the Conseil had the direction, generally speaking, of public hygiene. The Conseil de Saluhrite of the Seine in its later development was composed of fifteen titular and six supplementary members, including also several honorary members, with others, who, by virtue of their oflace, were members of the committee. These were, the Dean, the Professors of Hygiene, of Legal Medicine, and of the Faculty of Medicine, a Member of the " Conseil de Sante des A7'mees," the Director of the School of Pharmacy, the General Secretary of the Prefecture of Police, the Inspector-General of the bridges and causeways, besides engineers, architects, and the chiefs of the police departments. Most of the provinces followed the example of the capital, and established " Conseils de Scdu- brite." All these boards, whether provincial or Pai^isian, had one essential feature in common — viz., that the medical, veterinary, and chemical professions were always represented on them. Whatever expert a town possessed, would probably have a voice, and find a seat, in the ^^ Conseil de Saluhrite." From these health boards, or committees, very excellent reports have emanated, and they continue at the present time to do useful work. III.— adulteratio:n" in gePcMany. § 13. If we turn to the records of Germany, we find that all those who adulterated foods or drinks in the Middle Ages were punished severely, with painful and dishonouring penalties, such IG FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 14, 15. as public exposure of the fraud and whipping at the gate. The earliest regulations* related more to the goodness of the work and the general quality of the goods produced, than to adulteration. Eveiy considerable trade was a little coi*poration,andbad workman- ship or falsity in the goods offered was an oSence against the guild itself; the member was consequently expelled or punished by the officers of the guild. For example, in 1272 the two sworn masters of the bakers' guild at Berlin were held responsible for seeing that good bread was baked. The tailors of Berlin and the bakers of Basil excluded a man for ever from their respective guilds, if guilty of bad workmanship (1333). The Berlin weavers, not quite so severe, excommunicated the offender against their regu- lations for one year (1295). The false butcher at Augsburg (1276) was expelled from the city for a month. In Nuremberg almost everything was regularly inspected ; there was a Backer- schcm, a " Safranschau," and a " Schau" with regard to brandy, drugs, syrup, hops, roses, tobacco, iron, meat, salt-fish, honey, leather, and many other things. It was at a " Safranschau " in 1444, that one Jobst Fendeker was burnt, together with his false saflVon, and in the following year two men and a woman were buried alive there for the same offence, f In all the cities of Germany there were copious regulations ■with regard to three things — bread, wine, and drugs. § 14. Bread. — In Nuremberg, in the fifteenth century* the baker was not allowed to mix the different kinds of corn, which must be baked separately. In Augsburg, it would appear that there were no less than six different kinds of bread. | The punisliments for offending bakers wei-e various. In some places the delinquent was put in a basket at the end of a long pole and ducked in a muddy pool, similar treatment to that which in England befell '■'■ the Scolds." § 15. Wine. — According to an old Augsburg chronicle, it was in 1453 that the adulterated wine of the Franks first appeared in that city ; but there is abundant evidence to show that wine had been tampered with previously, and in 1390, one Ludwig von Langenhaus was sentenced to be led out of the city with his iiands * A small work, " Der Kampf gegen die Lebensmittelfalscliung von Ausgang des Mittelalters zum Ende des 18. Jahrhunderts, von L. Wasser- niann, Mainz, 1S79," contains some very interesting particulars with regard to the regulations in practice in the Middle Ages, both as to General Hygiene and the Adulteration of Food. t Henry II. of France enacted that if safifron was adulterated, the offenders should be punished by corporal chastisement, the drugs confiscated and burnt. t Maurer, " Geschichte der Stadtverfassung in Deutschland." Bd. III., a. 24. § 15.] EARLY ADULTERATION IN GERMANY. 17 bound and a rope round his neck, because of his practices in the adulteration of wine. In 1400, two wine-sellers were branded and otherwise severely punished, and about the same period a special law enacted forbidding the sulphuring of casks, the colouring of wine, or the addition of sugar, honey, or other sweet things. In the year 1435, says the old chronicle, " were the taverner Christian Corper and his wife put on a cask in which he had sold false wine, and then exposed in the pillory. The punish- ment was adjudged because they had roasted pears, and put them, into new sour wine, in order to sweeten the wine. Some pears were hung round their necks like unto a Paternoster." It further appears that they narrowly escaped being burnt. In 1451, the city of Cologne made a strong repi'esentation to the governing body at Antwerp on the prevalent adulterations of wine. At Biebrich on the Ehine, in 1482, a falsifier of wine was condemned to drink six quarts of his own wine. He died from the effects.* In the fifteenth century at Ulm, every tavern-keeper had to ajjpear at stated times before the sheriff" (Stadtrechnerj) and swear that neither he, his wife, his ohildx-en, nor any one else in his name, had mixed with the wine, woad or extract of woad, chalk, mustard seed, clay, " Scharlachkraut," must of apples, lead, mercury or vitriol ; no water might be added, and the same wine was to be retailed as bought. The Stadtrechner had also to see that no sour, ropy, or otherwise bad wine was sold. In the same town an Onlonnance of 1499 decreed, that since adulteration was most readily practised by putting substances into the cask, no cask was henceforth to be closed up save by the sworn cooper, who on finding anything amiss, was to give information; penalty for default, a guilder. In the fourteenth century, Nuremberg was a great centre of the wine trade ; consequently in that city there were very many regulations against adulteration of wine, but they were similar to those already mentioned. At Frankfort on the Maine, on false wine being found, the cask was placed on the knacker's cart, and a red flag displayed, with the inscription " stummer Wein" that is, mute or dumb wine. The jailer marched before, the rabble after, and when they came to the river they broke the cask and tumbled the stuff into the stream. J About the same date, wine is said to * In some places in very early times, the regular legal penalty was capital punishment; for example, in 1269, the law of Ripen punished the seller of false honey and wax with death. t Perhaps " Stadtrechner " might be translated " Mayor." X The article for sale was sometimes merely forbidden to be sold : this was especially the case with importations from other countries. For example, at Cologne, in 14S3, casks of butter imported from England, which were found o 18 FOODS: THEIR COMPOSITION AND ANALYSIS. [§ 16, 17. Lave been sophisticated with the following substances : earth, eggs, albumen, argol, mustard, salt, burnt salt, sweet milk, brandy, almond milk, wheat flour, clay, and several other things. In the early part of the seventeenth century, a circumstance happened in AVurtemberg, which led to some stringent regiila- tions with regard to metallic contamination of wine. A number of people having been seized with colic, paralysis, and other symptoms of poisoning by lead, it was noticed that all those attacked drank -one particular species of wine only ; and on investigation the epidemic Avas discovered to be due to the contamination of the wine by the use of metallic fastenings to the casks. The occurrence is related by Gockelius,* who styles the disease " Weinkraoikheit." § 16. Drugs. — Those who sold drugs, roots, spices, and the like, were strictly supervised, and in the reign of Frederick the Second of Prussia the examination of drugs was made a special calling, the inspectors being appointed by the king. In Augsburg, Frankfort, and a few other places, the ti-ade in medicine was taxed. By virtue of this tax the druggists and doctors enjoyed a monopoly, and medicines were forbidden to be sold by other trades. In the seventeenth century there were committees of doctors,, whose duty it was to inspect the druggists, and from the com- mittees — as in London, so on the Continent — originated the pharmacopoeias. Thus Pharmacopoeia, Antwerpiensis, 16G1 ; P. Utrajectina, 16G4; P. Amstelodamensis, 1668; Antidotariuvi Bun- oniense, 1674 ; liegia Chemica et Galenica, Geneva, 1684, &c.t lY.— HISTORY OF ENGLISH LEGISLATION IN EEFERENCE TO THE ADULTERATION OF FOOD. § 17. The first General Act in this country was the Act of 1860, previous to which date individual articles, such as tea, coffee, chicory, beer, and wine, were legislated for by special statutes, the to contain a mixture of old aud new, were not allowed to enter the market. In like manner false oil was excluded from the city. * "Beschreibung der Weinkrankheit," 1637. t The dukes of iSaxony regulated the trade of druggist as early as 1607. J. Guillaume published Recjitments entre les midecins et les apothicaires pour la visite des drcxjues. Dijon, 1605. Thomas Bartolin edited the work of Licetti Benanci—i)f c/nr«^4o Jraudum quae apud pharmacopeos commit- tuntiir. Franc. 1667 and 1671, 8vo. § IS.] LEGISLATIOX OX THE ADULTERATION OF FOOD. 19 object of wliicli was, for the most pai't, to j)revent the defrauding of the revenue; the healtli of the purchaser, and the injury done to him, being somewhat less considered, although not lost sight of. § 18. Bakers — llrmd. — An Act in the reign of George IV. was directed particularly againt the use of alum. Bakers, either journeymen or masters, using alum, and convicted on their own confession or on the oath of one or two witnesses, were to forfeit a sum not exceeding £20, and not less than £5 ; if beyond the environs of London, a sum not exceeding £10, nor less than £3. If within London or its environs, the justices were allowed to publish the names of the offenders. The adulteration of meal or dour was punishable by a like penalty, and loaves made of any other grain than wheat were to be distinguished by a large M. The possession on his premises by a milkn-, baker, &c., of any ingredient adjudged to have been placed there for the pur- poses of adulteration, was ])unishable by fine. An Act passed in 1836 (G and 7 Will. IV., c. 37) relative to bi-ead, may be considered as a modern development of the old " Assize of Bread." It repealed the several Acts relating to bread sold out of tlie city of London and the liberties thereof, and beyond the weekly bills of mortality and ten miles from the Royal Exchange, and provided other regulations for the making and sale of bread, and for preventing the advilteration of meal, flour, and bread beyond the limits aforesaid. This Act made it lawful for the bakers to make bread of " wheaten llour, barley, rye, oats, buckwheat, Indian corn, pease, beans, rice or potatoes, or any of them, and with any common salt, pure water, eggs, milk, barm, leaven, potato or other yeast, and mixed in such pro- portions as they think lit, and with no otlier ingredient or matter whatever." They were also permitted to make the bread any weight or size they chose. The bread, however (sect. 4), was to be sold by weight, and in no other manner. Section 3 provided that no baker or other person within the limits prescribed . . . " shall use any mixture or ingredient whatever in the making of such bread other than, and except as hereinbefore mentioned, on any account or under any colour or pretence whatsoever ; " penalty not less than £5 and not more than £10. In default of payment the offender could be im- prisoned for a period of not more than six months, with ov without hard labour,* or unless the penalty was sooner paid, the magistrates could also i-ublish the offender's name, and defray the expense of such publishing out of the fines. * In the case of Cobe v. James, 41. L. J. M. C. 19, it was held that for a conviction under this section guilty knowledge was necessary. 20 foods: their composition and analysis. [§19,20. By section 9 no person is to put into any corn, meal, or flour, at the time of grinding, dressing, or manufacturing any ingredient or mixture whatsoever, not being the real and genuine produce of the corn or gi-ain ; nor is flour of one sort or corn to be sold as flour of another sort. Penalty not less than £5 and not more than £20. Sections 11 and 12 provide that bread made partially or ■wholly of pease, beans, or potatoes, or of any other sort of corn or gi'ain other than wheat, is to be marked with a large Roman M, under a penalty not exceeding 10s. Justices may issue a search warrant to enter a baker's premises at reasonable hours, and search for adulterated bread or substances used for adul- teration. If such substances have been found, the justices may dispose of them as they think fit. The penalty for a first off"ence is a fine not exceeding <£10 and not less than £2 ; for the second oflence £5, and for every subsequent off"ence £10. In default of payment, imprisonment for not more than six calendar months. It is also lawful for the magistrates to publish the name of the offender. The Act contains various other provi- .sions with regard to obstructing search, offences committed by a journeyman or servant, the baking of bread on the Lord's Day, (tc. § 19. Bee7' and Porter. — An Act, passed in 1816, 56 Geo. III., c. 58, enacted that no " brewer of, or dealer in, or retailer of beer," sliall receive or have in his possession, or use or mix with any worts or beer any molasses, honey, liquorice, vitriol, quassia, cocculus indiae, gi'aius of paradise, guinea, pepper or opium, or any article or preparation whatever for or as a substitute for malt or hops. If any person contravene this provision, the officer of the excise may seize the worts and beer, together with the casks con- taining the same." — Penalty £200. By the same Act, druggists who sold colouring materials or malt substitutes to brewers, were liable to a penalty of £500. An Act, 7 and 8 Geo. lY., c. 52, pro- vided that any brewer, having in possession or in use, substitutes for malt or hops, or for darkening the colour of beer, was liable to a penalty of £200, and the ingredients, beer, worts, &c., might be seized. Later on, by the 10 and 11 Vic, c. 5, brewers were allowed to make for their own use a colouring-matter out of sugar.* § 20. Wines.— An old statute (51 Hen. ill., st. 6) forbade the use of unwholesome wine or meat. By the statute 12 Chai-les II., c. 25, any adulteration of wins was made punishable, with the forfeiture of £100 if done by the wholesale merchant, and £40 if done by the vintner or retail trader. Additional regulations * At the present time, it is not illegal to mix bitter beer with whole- some bitters other than hops. § 21, 22.] LEGISLATION ON THE ADULTERATION OF FOOD. 21 ■were made by I. William and Mary. All of these Acts, it is scarcely necessary to say, are now obsolete and repealed. § 21. l^ea. —An Act was passed in 1723, 11 Geo. I., c. 30, which enacted that " no dealer in tea, or manufacturer or seller thereof, or pretending to be, shall counterfeit tea, or adulterate tea, or cause or procure the same to be counterfeited or adulterated, or shall either fabricate or manufacture tea with terra japonica, or with any drug or drugs whatsoever, nor shall mix, or cause or procure to be mixed with tea any leaves other than leaves of tea, or other ingredients whatsoever, on pain of forfeiting and losing the tea so counterfeited, adultei-ated, altered, fabricated, manufactured, or mixed, or any other thing or things whatso- ever added thereto, or mixed or used therewith, and also the sum of £100." Some years afterwards, 1730-31, afiirther Act was passed [4Geo. II., c. 14] prescribing a penalty for what is termed in the statute "sophisticating tea." It recites "that several ill-disposed persons do frequently dry, fixbricate, or manufecture very great quantities of sloe leaves, liquorice leaves, and the leaves of tea that have been before used, or the leaves of other trees, shrubs, or plants in imitation of tea, and do likewise mix, colour, stain, or dye such leaves, and likewise mix tea with terra japonica, sugar, molasses, clay, logwood, and with other ingredients, and do sell and vend the same as true and real tea, to the prejudice of the h.ealth of His Majesty's subjects, the diminution of the revenue, and to the ruin of the fair trader.'' The penalty under this statute was XIO for every pound of tea sophisticated. The next Act was passed in 1776, 17 Geo. III., c. 29. The preamble asserts that great quantities of sloe leaves and the leaves of the ash, elder, and other trees, shrubs, and plants, were manufactured in imitation of tea, and were then sold to dealers in tea, who, after mixing the leaves with tea, sold it as time and real tea ; but as the persons who fabricated or manufactured the leaves were not dealers in tea, they were not punishable by the law then in force. The Act, therefore, rendered any person, whether a dealer or not, who fabricated leaves in imitation of tea, or who mixed tea with other ingredients, or who sold, exposed for sale, or had in his custody fabricated or mixed teas, liable on conviction to a penalty of £5 for each pound of such tea, or in default to imprisonment for not less than six months, nor more than twelve. The officers of the excise were empowei-ed to enter under warrant any premises by day or night and seize the leaves, which, on a further warrant, were to be destroyed. § 22. Coffee. — Tlie history of the regulations with regard to coffee and chicory is rather curious, inasmuch as coffee appears 23 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 22 to have been adulterated almost immediately after its introduc- tion, and legislative interference was soon necessary. The first Act, 5 Geo. I., c. 11, 1718, with regard to coffee, recited tliat " Whereas divers evil-disposed persons have at the time of, or soon after, the roasting of coff"ee made use of water, grease, butter, or such like materials, whereby the same is rendered unwholesome, and gi'eatly increased in weight to the prejudice of His Majesty's revenue, the health of his subjects, and the loss of all honest and fair dealers in that commodity," and went on to enact that " any person or persons whatsoever, who shall at the roasting of any coffee, or before, or at any time afterwards, make use of water, grease, butter, or any other material whatsoever, which shall increase the weight, or damnify or prejudice the said coffee in its goodness, he, she, or they shall forfeit the sum of £20 for every sucli offence." This penalty was inci-eased to £100 by a subsequent Act (12 Geo. I., c. 30) passed in the vear 1724. An Act passed in 1803 (43 Geo. III., c. 129), ordered that "If any burnt, scorched, or roasted peas, beans, or other grains, or vegetable substance or substances be used, prepared, or manu- factured for the pui-pose of being an imitation of, or in any respect to i-esemble coffee or cocoa, or to serve as a substitute for coffee or cocoa, or alleged or i)retended by the possessor or vendor thereof so to be, shall be made or kept for sale, or shall be found in the custody or jjossession of any dealer or dealers of coffee or cocoa; or any burnt, scorched, or roasted peas, beans, cfec, not being coffee or cocoa, shall be called by its preparer or manu- facturer, possessor, or vendor thereof, by the name of coffee, or by the name of American cocoa, or English or British cocoa, or any other name of cocoa, the same respectively shall be for- feited, and togetlier with packages containing the same, shall or may be seized by the excise." The person convicted was to be fined £100. A subsequent Act (3 Geo. IV., c. 53) permitted j)er:ions, not dealers in coffee, to sell roasted corn, peas, beans, or parsnips whole, but not ground, crushed, or powdered. In 1832, grocers were allowed by the Excise to keep chicory on their premises, and a Treasury minute, dated August 4, 1840, allowed the sale of coffee mixed with chicory, a step which no doubt opened tlie way to wholesale adulteration. This is evident from a meeting of those interested in the coffee trade, held at the London Tavei-n, ou the 10th of March, 1851, in which the chairman* explained that althougli more of what was called coff"ee was now consumed, yet that there was a less consumption *T. Baling, Esq., M. P. § 22.] LEGISLATION ON THE ADULTERATION OP FOOD. 23 of genuine coffee. " We wish," he continued, " to come to the real question, and we desire that it should be publicly under- stood that what is coffee be sold as cotfee, and that what is not coffee, being a cheaper article, and, if you will, a more nutritious article, and as eligible for consumption, be sold to the consumer at the price at which it can be afforded." A grocer from Shore- ditch having produced at the meeting a compound of burnt peas, dog-biscuit, prepared earth, and a substance "which," he said, ■" I shall not describe, because it is too horrid to mention," went on to affirm that several tons of the same material were in existence, and that it was used as a substitute for chicory and for snuff. * The Lancet also gave details about the same time of the micro- scopical examination of thirty-four samples of chicory, nearly one-half of which were mixtui-es, the substances found being roasted beans, burnt corn, and acorns. It was under the pro- tection of this Minute, that Messrs. Duckworth of Liverpool took out a patent for the compression of mixtures of chicory and coffee into the shape of berries. Popular writers have, as usual, made the most of this patent, and the story has been i-etailed with additions from one book to another as a glaring instance of whole- sale fraud ; but, although the purity of the manufactui'ers' inten- tions may be open to doubt, the ftvct remains that they did nothing against the existing law. The patent does not appear to have been profitable, and but few of the chicory berries wex*e put in circulation. The subject of coffee-adulteration was not, however, per- mitted to escape tlie attention of Parliament, and petitions from planters, growers, and others interested in the sale or production of coffee increased in number. In the Commons, during the course of a long debate (April 14, 1851), Mr. T. Baring stated that it could not be denied that there had been a diminu- tion since 1847 in the consumption of coffee to the extent of six million pounds, the real cause of which was the wholesale ixdmixture of coffee with chicory — this chicory of home growth. In 1840, at the time of the issuing of the Treasury-minute sus- pending the law as regarded that article, all the chicory used in the country came from abroad, and as an excisable import on which duty was paid, but since the issue of the Minute it had been cultivated largely in England. Similar statements were made in the House of Lords (July 3, 1851), on the occasion of Lord Wharncliffe's presenting a peti- * " Adulteration of Coffee. A verbatim report of the proceeduags at a public meeting held at the London Tavern. " London, 1851, . 24 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 23, tion to that House. In the following year, under the pressure of popular feeling on the subject, the objectionable Minute was rescinded, and a new Treasury-minute, dated July 27, 1852 (which was afterwards embodied in a subsequent order of the Inland Eevenue Commissioners), permitted licensed dealei'S in coiTee to "keep and sell chicory and other vegetable substances prepared to resemble coffee, provided that they be sold unmixed with coffee, in packages sealed or otherwise secured, containing respec- tively not less than two ounces, and having pasted thereon a printed label with the name of the seller, the exact weight and true description of the article contained therein, and provided that no such article be kept in a loose state, or otherwise than in such packages as aforesaid, in any room entered for the storage or sale of coffee." This regulation was, without doubt, irksome both to traders and consumers, since every one who desired his coffee mixed with chicory could not buy the mixture prepared, but was obliged to purchase the coffee and chicory sepai'ately, and compound it himself Hence, many memorials were presented l)raying " Tliat the sale of a mixture of coffee and chicory be not interfered with, provided each package has legibly printed thereon words plainly indicating such mixture." In consequence of these representations this Minute was also rescinded* and a new one prepared. An order of the Commissioners of Inland Eevenue, dated May 13, 1853,t followed, requiring that on every package containing a mixture the words "This is sold as a mixture of chicory and coffee," be printed in capital letters of Roman char- acter, of at least one-eighth of an inch in height, on the outside of the packages or canisters, on the same side of which there was to be no other printing or writing. On no other part of the pack- age, further, was there to be any other writing save the name and address of the seller. It would be a great error to suppose that these minutes of the Treasury, and subsequent orders of the Revenue Commissioners, had for their leading object the prevention of adulteration in its reference to the health of the subject. It will at once be noticed that they only touched on " excisable articles," and it was entirely a fiscal question. In a word, had the sophistication been of such a nature as to increase instead of diminish the revenue, the Treasuiy would have let it pass without notice. § 23. The Select Committee of 1855. — The prelude to legislation on adulteration as a whole, was the appointment of " A Select Parliamentary Committee," which entered on its labours in 185&. * Pari. Paper, No. 165, Vol. xcix., sess. 1852 53. t Pari. Paper, Ko. 508, sess. 1854-55. § 23.] LEGISLATION ON THE ADULTERATION CF FOOD. 25 The early appointment of this committee was, without doubt, clue to the influence of the late Mi'. Wakley, the able and courageous editor of the Lancet. In 1850 Mr. Wakley had established, in connection with his powerful journal, " Tlie Lancet Sanitary Com- mission," of which commission Dr. Hassallwas the leading spirit, with Dr. Letheby as occasional coadjutor in matters purely chemi- cal, and (what at that time was of great importance) with the assistance of an artist, who drew microscopical objects with fidelity. The "Analytical Sanitary Commission" was commenced in the first number of the Lancet for 1851, and the scope of the inquirj% as stated by the editor, was as follows : " We pro])ose, for the public benefit, to institute an extensive and somewhat rigorous series of investigations into the present condition of the various articles of diet supplied to the inhabitants of this great metropolis and its vicinity. . . . Special features of the inquiry will be that they are all based upon actual inquiry and experiment ; the microscope and the test tube will be our constant companions." Notice was also given that at the expiration of three months the names and addresses of the shopkeepers from whom purchases had been made would be given; but at the commencement the sti'eet alone was to be indicated. The promise was kept, and hazardous although the experiment most certainly was, yet in April we find the names of large firms freely published, and, so to speak, "pilloried," for having sold impure and false goods. In 1855 Dr. Hassall collected the articles which had been pub- lished in the Lancet into a volume, entitled " Food and its Adul- terations, comprising the reports of the Analytical Sanitary Commission of the Lancet for the years 1851-54. London, 1855." In 1855 " The Select Committee on the Adulteration of Food" commenced its labours, and examined as far as possible all those who were likely to have any special knowledge of the adultera- tions themselves, the methods necessary to detect them, and their eflfect on the revenue and on health. Dr. Hassall stated to the committee the results of his inquiries both for the Lancet Cora- mission and during the course of his other labours, and gave in detail the frauds practised in regard to milk, cofiee, tea, drugs, preserved fruits, &c. Dr. Alphonse Normandy, who had also written a work on adulteration — the result of ten years' investigation — said, in giving evidence as to the aluming of bread, that he had seen alum in bread in crystals of the size of a large pea. " In the bread of one baker I found alum actually in the state of large crystals ; I went to him and showed him his bread, and he said, ' I cannot help it.' " In extreme instances he had found as much as from 2G FOODS: THEIR COMPOSITION AND ANALYSIS. [§24. 250 grains to twice that quantity of alum in the 41b. loaf, and in 1847 he had found magnesia carbonate in three samples of bread. In 1847 and 1848, years of gi'eat scarcity, he had discovered bean and pea meal in flour, but this he considered quite exceptional. Witli regard to beer, he thought that brewers often made use of cocculus indicus ; and, finally, he gave evidence of the great adulteration of drugs. Mr. Blackwell, of the firm of Crosse & Blackwell, gave some very interesting evidence as to the " coppering of preserved vege- tables " practised before the food-articles appeared in the Lancet. The process in use by his firm was to boil the pickles or vinegar several times in copper boilers. After each operation they became greener, and when the proper hue was attained, tlie process was finished ; but since the outcry on coppered vegetables, this process had been abandoned. Another witness, Mr. 0. L. Simmonds, the author of a work upon " Commercial Products," in giving evidence on the adultera- tion of drugs, estimated that there was a loss to the revenue from this cause of no less than £3,000,000 per annum. As an instance of the manner in which the revenue sufi"ered, he cited the substi- tution of cassia for cinnamon ; cassia paid Id. per lb. duty, cinnamon 2d. The dealers sold cassia under the name of cinna- mon to such an extent as to aflect seriously the cinnamon trade. § 24. Adulteration Acts, 1860 and 1872. — Upon the report of tlie Select Committee, the first general Adulteration Act was drafted, and became law in 18G0. The first section enacted, " That every person who shall sell any ai-ticle of food or drink with which, to the knowledge of such person, any ingredient or material injurious to the health of persons eating or drinking such article, has been mixed, and every person who shall sell as pure or unadulterated any article of food or drink which is adulterated or not ]iui(', shall for every such offence, on summary conviction of the same, pay a penalty not exceeding £5, with costs." A second offence was punishable in addition by publishing the offender's name, place of abode and offence. The Act permitted, but did not compel, the appointment of analysts. The bodies which might appoint such analysts were — in the City of London, the Commissioners of Sewers ; in the metropolis generally, Vestries and District Boards ; in the counties, Courts of Quarter Sessions. Section 4 provided that any purchaser of any article of food in any of the districts in which analysts existed, miglit have such article analysed on payment of a sum not less than 2s. Gd. and not more than 10s. 6d. ; the purchaser, on the com- pletion of the analysis, was entitled to receive a certificate of the result of the analysis. § 24.] LEGISLATION ON TUE ADULTERATION OF FOOD. 27 These appointments were at first confirmed by the Secretary of State, but afterwards the Local Government Act of 1871 trans- ferred the regulation of the appointments to the Local Govern- ment Boai'd. The Act existed, and was in partial operation, for twelve years, when it was entirely recast and inter.speised with various sanitary considerations. In an Act passed in the year 1872 (35 and 3G Yic, c. 7'i), it was enacted that " Every person who shall wilfully admix, and every person who shall order any other person or persons to admix, any ingredient or material with any drug to adulterate the same for sale, shall be liable to a penalty for the first ofi'ence not exceeding £50, with costs." The second ofience was punish- able by a term of imprisonment not exceeding six months, with hard labour. By the second section, " Every person who shall sell any article of food or drink, with which to the knowledge of such person any ingredient or material injurious to the health of per.sons eatiug or drinking such article has been mixed, and every person who shall sell as unadulterated any article of food or drink or any drug which is adulterated, shall for every such oftence, on a summary conviction of the same, pay a penalty not exceeding o£20, with the costs of conviction." By the third section " Any person who shall sell any article of food or drink, or any drug, knowing the same to have been mixed with any other substance with intent to fraudulently increase its weight or bulk, and who shall not declare such admixture to any purchaser thereof before delivering the same and no otlier, shall be deemed to have sold an adulterated article of food or drink, or drug, as the case may be, under this Act." The Act, with doubtful advantage, also extended the right of ■appointing analysts to boroughs having separate police establish- ments. The appointment was optional, save on the direction of the Local Government Board. The sixth section provided that inspectors of nuisances or other local officers were to procure samples for analysis. Private purchasers might have articles analysed as before, the only difference being that, under this Act, they were to hand the substance, not to the analyst, but to the inspector. There were also provisions as to the sealing and division of samples. Since the Act of 1860 remained unrepealed, the two Acts wei-e both in force simultaneously, and under their joint operation the following off'ences were punishable : — * 1 . Selling any article of food, drink, or medicine, that contains any ingredient injurious to health, and knowing it to contain such ingredient. 2. Selling any adulterated food, drink, or drug. * "The Law of Adulteration," by Sidney Woolf. Lond. 1S74. 28 FOODS : THEIR composition and analysis. [§ 25. 3. Wilfully mixing with any article of food or drink any in- gredient or poisonous ingredient to adulterate the same for sale. 4. Wilfully mixing any ingredient with any drug to adulterate the same for sale. 5. Selling any article of food, drink, or any drug, knowing the same to have been mixed with other substances with intent frau- dulently to increase its weight or bulk, imless such admixture be declared at the time of sale. § 25. The Select Committee, 1874. — These Acts by no means worked well. Many of the analysts were inexperienced, and even those who had considerable chemical knowledge differed widely in the conclusions they drew from their analyses. The reason of this was evident, for the standards had scarcely been settled. There was, for example, no general agi-eement as to the amount of "fat" and "total solids" in milk ; the question of whether tea should be permitted to be faced, or not, Avas then (as, indeed, now) unsettled; there was no method in use which distingiiished alum added to flour and alumina existing as sand. Analyst contra- dicted analyst. Magistrates were perj^lexed as to the meaning of the word " adulteration," and conflicting decisions on mere legal technicalities offered a still further obstacle to the healthy opera- tion of the Act. The public generally were dissatisfied with an Act which on many retail dealers inflicted real hardships — e.g.^ tea, paid for at the highest market price, and imported direct from China, would be examined by a local analyst, and pronounced to be faced with Prussian blue, gypsum, A:c. ; while, from the pecviliar nature of the statute, the seller, however innocent of the fraud himself, could not defend the charge on anything like equal terms. Petitions, moderate in tone, came in from most of the large towns, and the Government decided to appoint another Select Committee. A large number of witnesses — tea merchants, tea brokers, tea retailers, butter merchants, cocoa and coffee manufacturers, milk sellers, bakei's, and analysts — were examined by this new Committee in 1874 ; and on their evidence a report was based, which stated that after having sat fourteen days, and examined fifty-seven witnesses, the Commissionei's had arrived at the unanimous conclusion that, while the Act had done much good, it had, at the same time, inflicted considerable injury, and enforced heavy and undeserved penalties upon some respectable tradesmen. "This appeai-s to have been mainly due to the want of a clear understanding as to what does, and as to what does not, constitute adulteration, and in some cases to the conflicting decisions and inexpexience of the analysts. Your Committee are, however, of opinion, that the Act itself is defective and needs amendment." § 26.] LEGISLATION ON THE ADULTERATION OF FOOD. 29 The report went on to say that the adoption of the Act had been by no means general, and in many cases where it had been applied, its operation was of a very restricted character; for, even with competent analysts, if inspectors were not appointed at the same time, the Act remained a dead letter. All the London vestries had made appointments, but in only twenty-six out of seventy-one boroughs, and thirty-four out of fifty-four coun- ties, were there at that date official analysts. The examination of tea was recommended to be made on importation by the Customs. The Committee did not consider that the exact proportion of mixtures need be stated on a label, and they wished to recoi'd that mixed mustard and prepared cocoa had been long manufactured at Deptford for the supply of the Navy. They recommended that small districts should be consolidated, and that, as a rule, the boroughs in a county should be united with the county for the purpose of appointing one analyst for the entire district; and they pointed out that the only way to secure " the services of I'eally efficient analysts is to offer them a fair remuneration, which can hardly be done without the union of several Local Authorities in one appointment." The Committee concluded their report by remarking that the public was " cheated " rather than " poisoned." § 26. Sale of Food and Drugs Act, 1875 and 1879.— On this report was based the Act of 1875, which is at the present moment, with its amendment of 1879, the existing law, and the full consideration of which will be reserved for another Section ; an early defect in the Act, however, may be at once alluded to, for it had not been long in operation before its action was almost entirely stopped by legal ingenuity. The sixth section provides that " No person shall sell to the prejudice of the purchaser any article of food or any drug which is not of the nature, substance, and quality of the article demanded by such purchaser;" and in a Justiciary Appeal case at Edinburgh, in which an inspector had purchased cream not for his own use but for analysis, the Scotch coui't discussed the "prejudice" question — three out of seven judges adopting the view that a purchase made under these con- ditions was not to the " prejudice" of the purchaser, and five out •of the seven dismissing the summons on other gx-ounds. The impression produced in this country, however, by the decision of the court, was that the sale, to be eflTectual, must be made iu the ordinai-y way, and not merely for the purposes of analysis. The same question was raised in quite a different but equally ingeni- ous way in a "mustard case" argued before the Court of Queen's Bench. The purchase in this case was by an officer ; the defence being that, as it was well known that mustard was mixed with 30 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 2G. flour and other things, sucli a purchase could not be to the pre- judice of the ])urchaser. Tlie point, however, was left undecided ; the question again came before the Court of Queen's Bench, in the case of Sandys v. Small, and the "prejudice" question was- argued on both points. Whisky was alleged to have been mixed with water, and the defence set up — that it was known to be so- mixed, and therefoi-e not to the prejudice of the purchaser — was held by the court to be good, and the case having been decided upon this point, the other was not proceeded with. Finally, the question was settled by the case of Hoyle t\ Hitchman, March 27, 1879. The facts in this instance were of the simplest char- acter: the appellant had purchased milk in the usual official way; the milk was found to be adulterated, and the defence was that, as he did not use the milk, therefore he did not })ay the milk for his own use; he was not prejudiced. The magis- trate who heard the case considered the defence good, and dis- missed the summons. Justice JMellor, in giving judgment, observed that the "pi'eju- dice" view of the Act "would absolutely nullify its beneficial effect. For if the meaning of the enactment is that the offence cannot be complete without its being 'to the prejudice of the purchaser,' it is hardly possible that the offence should be brought home to any one. And this observation, in my view, goes far to show tliat this construction cannot be the right one. So far as authority is concerned, there is no direct decision in favour of such a view ; and indeed, in the English courts there is hardly any authority upon the point. For in the first of tlie two cases in this court referred to, the mustard case, my brother Lush distinctly said that, in his view, if the article were adultei'ated, it must be presumed that it was ' to the prejudice of the pur- chaser,' and I could not have dissented from that opinion, or I could not have concurred in sending the case down to be re-stated on the other point. And as to the other case, no doubt in the course of the argument the Lord Chief-Justice made some such remark, but not by way of a decided dictum, and rather by way of query or suggestion, and the decision went upon the other point, so that there is no authority in the English courts in favour of the view now presented. It cannot be said that the weight of judicial authority is against, and I rather tliink it is in favour of, the view which we have arrived at after the best consideration given to the question, as to the true construction of the enactment. It is quite general in its terms, and its terms are very large, nor is there anything to limit them, — 'if any one shall sell, to the prejudice of the purchaser, any article of food not of the nature, substance, or quality of the article demanded by the purchaser.* § 20.] LEGISLATION ON THE ADULTERATION OF FOOD. ol There is nothing to limit the application of the enactment (as some of the Scotch judges seem to have supposed) to articles deleterious in their nature. And in several of the sections (13 to 17) provisions are made for purchases by public officers for the purpose of analysis and prosecution, assuming that if the article is found to be adulterated, the offence will have been committed. It would be strange indeed if all these provisions were to be made nugatory by a construction which would, in effect, come to this — that proceedings could only be taken by private individuals. Here the purchase was made by the inspector nnder those sec- tions ; but surely the case must be treated as though the purchase had been by a private individual. Now, in the case of a private individual no one could dispute that in such a case as this the off"ence would have been completed, and the magistrate has so found, in fact. That being so, what difference can it make as to- the nature of the offence, that the purchase was by an officer on behalf of the public, and furnished with public money for the jiur- pose ? If tlie purchaser asks for a certain article, and gets an article which by reason of some admixture of a foreign article is not of the nature or quality of the article he asks for, he is neces- sarily 'prejudiced;' and how can the fact that the purchase is not with his own money at all aff"ect the question of the commis- sion of the oflfence? The offence intended to be prevented by the Act was the fraudulent sale of articles adulterated by the admix- ture of foreign substances, which would necessarily be ' to the prejudice of the purchaser;' and those words were inserted only to require that such an adulteration should be shown to have been made. Taking all these matters into consideration, I cannot bring my mind to the conclusion that in such a case as this the offence is less complete, merely because the money with wliich the purchase was made was not the money of the purchaser, which must be wholly immaterial to the seller, and cannot affect the offence he has committed. I come, therefore, to the conclusion that the magistrate was wi-ong in dismissing the case on that ground, and, therefore, that the case must be remitted to him to be determined on the evidence as to the offence alleged to have been committed." Mr. Justice Lush, in expressing his entire concurrence, said that the diflferences of opinion which unfortunately prevailed as to the true construction of the sixth section of the Act had crip- pled the operation of a most beneficial Act. — Judgment for the appellant. Finally, the Act of 1875 was amended by the " Sale of Food and Drugs Act, 1879," which became law on July 21st in that year. This Act settled the "prejudice" question, authorised the 32 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 27, 28. obtaining of samples of milk for the purposes of analysis, and established standards for spirits. (See sections on the "Existing Law relative to Adulteration.") v.— THE HISTORY OF THE PRESENT SCIENTIFIO PROCESSES FOR THE DETECTION OF ADULTERATION. § 27. If an attempt were made to write the full history of the modern system of the practical assaying of foods, beverages, and drugs, the result would be neither more nor less than a history of the development of the chemical, physical, botanical, and medi- cal sciences ; for there has scarcely been a single advance in any one of those sciences which has not some bearing, immediate or remote, on our subject. Hence, the more useful and less am- bitious method to pursue will be merely to notice the chief writ- ings and the more noteworthy discoveries of those who have explored this special field of investigation. The very early and brief notices in the old writers have been already mentioned. The first general works on adultei'ation were devoted to drugs rather than to foods, and the herbals and the older works contain here and there, scattered through their prolix pages, casual mention of substitutions or fiilsifications. For example, Saladin of Ascala, a ])hysician to the Grand Constable of Naples, who wrote in the fifteenth century a work on the aromatic principles of drugs, describes methods of preserving food, and in speaking of the adulteration of manna with sugar and starch, cites the case of an apothecary who was fined heavily and deprived of his civil rights.* § 28. In the early part of the seventeenth century Bartoletus discovered by analysis milk-sugar (see chapter on "Milk"), and to this epoch belong also some observations and experiments of another Italian, San Francesco Redif of Florence, published in 1660, on the amount of mineral substances in pepper, ginger, and * This work, "Compendium Aromatarium," was published in Augsburg, 1481. There is no separate copy in the British Museum, but it will be found as the "Liber Saladini" in the beautiful folio edition of the Arabic phy- sician (Ydmanna ibn Massawaih), Joaunis mesuae damasccni medico claris- ■shni opera, histications des medi- 44 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 38. caments." Traduction par J. B. Kapeler et J. B. Caventou, Paris, 1821, 2 Vols., in 8vo. Branch:, Giuseppe. — " Sulla falsificazione delle sostanze specialmente medi- cinali e sui mezzi atti ad scoprirJi." Piza, 1823. Desmarest. — " Traite des falsifications, ou expose des diverses manieres de constater la purete des substances employees en mt5decine, dans les arts, et dans I'economie domestique." Paris, 1827, in I2mo. BUSSY ET Boutrox-Charlard. — "Traite des moyens de reconnaitre les falsifications des drogues, simples et composees, et d'en constater le degre de purete'." Paris, 1829, Sv^o. Walchner, F. H. — " Darstellung der wichtigsten im blirgerlicLen Leben vorkommenden Verf'alschungen der Nahrungsmittel und Getriinke, nebst den Angaben, wie dieselben schnell und sicher entdeckt werden kouneu." Karlsruhe, 1840, iu 12mo, 120 pp. . — "Darstellung der wichtigsten, bis jetzt erkannten Ver- fUlschungen der Arzneiniittel und Droguen." Karlsruhe, 1841, 8vo, Brum, Franz. — "Hilfsbuch bei Untersuchungen der Nahrungsmittel und Getranke, wie deren Echtheit erkannt und ihre Verfalschungen entdeckt werden kbnnen." "Wien, 1842. RlCHTER. — " Die Verfalschungen der Nahrungsmittel und anderer Lebens- bediirfnisse, nebst einer deutlichen Anweisung die Echtheit derselben erkennen und ihre Verfalschung entdecken zu kounen." Gotha, 1843. Garnier, J., ET Harel, Cn. — "Des falsifications des substances alimen- taires, et des moyens chimiques de les reconnaitre." Paris, 1844. Bertin, G. — " Sophistication des substances alimentaires, et moyens de les reconnaitre." Nantes, 1846, 8vo. Friedrich, J. B. — "Handbuch der Gesundheitspolizei, der Speisen, der Getranke, und der zu ihrer Bereitung gebrauchlichsten Ingredienten." Ansbach, 184G, 8vo. Beck, Lewis C. — " Adulterations of Various Substances used in Medicine, and the Means of Detecting them : intended as a Manual for the Physician, the Apothecary, and the Artisan." New York, 1847, 8vo. AcAM, F. L. — "Traite des falsifications des substances medicamenteuses et alimentaires, et les moyens de les reconnaitre." Anvers, 1848, in 8vo. Mitchell, J. — "Treatise on the Adulteration of Food." London, 1848, in 12mo. Pedroni, p. M. — " Manuel complet des falsifications des drogues, simples et composdes." Paris, 1848, in 18mo. Normandy, Alphonse. — "Commercial Handbook of Chemical Analysis." London, 1850 (there are later editions). Cheyallter, A.^ — " Dictionnaire des alterations et falsifications des sub- stances alimentaires, medicamenteuses et commerciales, avec I'indica- tion des moyens pour les reconnaitre." Paris, 1850-52, 2 Vols. (There are later editions.) Dungerville, Emile. — "Trait(^ des "falsifications des substances alimen- taires, et des moyens de les reconnaitre." Paris, 1850. Tauber, Isidore. — " Verfalschungen der Nahrungstofie und Arzneimittel. " Wien, 1851, 8vo. Pierce. — "Examination of Drugs, Medicines, Chemicals, &c., as to their Purity and Adulterations." Cambridge, Massachusetts, U.S., 1852, in 12mo. Gille, N. — " Falsifications des substances alimentaires." Paris, 1853, Fop. — " Adulteration of Food." London, 1858. § 38.] BIBLIOGRAPHY. 45 How. — " Adulteration of Food and Drinks." London, 1S55, Hassall, Arthur Hill. — " Food and its Adulterations, comprising the Reports of the Analytical Sanitary Commission of the Lancet for the years 1851-54." (There is a later edition.) HuREAUX. — " Histoire des falsifications des substances alimentaires et niedicamenteuses." Paris, 1855, 8vo. Marcet. — "Composition, Adulteration, and Analysis of Foods." London, 1850. Payen. — " Des substances alimentaires." Paris, 1856. Dalton. — " Adulteration of Food." London, 1857. SouiLLiER, J. — " Des substances alimentaires, de leur qualite, de leur falsification, de leur manutention, et de leur conservation." Anvers, 1858, 8vo. Klencke. — "Die Verfiilschung der Nahrungsmittel, Getranke," &c. Leipzig, 1858. Petit Lafitte. — "Instruction simplifii^e pour la constatation des propri^t^s des alterations et des falsifications, des principales denrees alimen- taires." Bordeaux, 1858. Gellee, a.- — " Precis d'analyse pour la recherche des alterations et falsi- fications de produits cliimiques et pharmaceutiques," Paris, 1860. Facen, Aurelio. — "Chimica bromatologica ossia guida per riconoscere la bonta, le alterazioni ' e le falsificazione delle sostanze alimentari. " Firenze, 1872. Walchner, J. H. — ' 'Die Nahrungsmittel des Menschen, ihre Verfiilschungen und Verunreinigung." Berlin, 1875. An Svo. of 324 pages, in which there are no plates and nothing new is advanced ; it is, however, a not unskilful compilation. Soubeiran, J. Leon. — " Nouveau dictionnaire des falsifications et des alterations des aliments, des medicaments," &c. Paris, 1874. The figures are mostlj' borrowed from Hassall's M^ork, the articles well compiled, French sources predominating, with very scanty notices of German work. At the end of each article there is a short biblio- graphy. Blyth, a. Wynter, — "A Dictionary of Hygifene, comprising the Detection of Adulteration." London, 1876. Hassall, A. H. — " Food and its Adulterations, and the Methods for their Detection." London, 1876. Naquet, a. — "Legal Chemistry; a Guide to the Detection of Poisons, Examination of Stains, &c., as Applied to Chemical Jurisprudence." New York, 1876. The work is a translation from the French, but superior to the original. At the end is a useful bibliography of works relating to forensic medicine and adulteration. Selmi, Antonio. — "Chimica applicata all'igiene, all' economia domestica." Milan. Klencke, Hermann.— "Illustrirtes Lexicon der Verfiilschungen der Nah- rungsmittel und Getranke." Leipsic, 1878. Sharples, C. H. — "Food and its Adulteration." Preston, U.S., 1879. Blane. — "De la Contrefa^on." Elsner, F. — "Die Praxis der Nahrunq-smittel, Chemikers Anleitung zur Untersuchung von Nahrungsmitteln u. Verbrauchsgegenstlinden, sowie f. hygienischen Zweck." Leipsic, 1880. Bell, James. — "The Analysis and Adulteration of Food." Pt. L, ISSl. Pt. II., 1883. London. 46 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 39. DiETZSCH, 0.— "Die Wichtigsten Nahrungsmittel u. Getriinke, deren Verunreiniguugen IX. Verfiilschungen." Zurich, 1 884. Elsner, F. — "Nahrungs- u. Genussmittel aus clem Pflauzenreiche, sowie deren Surrogate." 149 Microphotos., 10 plates. 4to. Halle, 1885. HiLGER, A. — " V^ereinbarungen betreffs der Untersixchung u. Beurthei- lungen von Nahrungs- u. Genussmitteln." Berlin, 1885. BiRNBAUM u. Grimm. — "Atlas v. Photographien Mikroskopischer Praparate der reinen u. gefalschten Nahruugsmittel." Vol. I. Atlas zur Mehlpriifg. IG plates. Folio. Stuttgart, 1SS6. Dammer, Dr. Otto. — Illustrirtes Lexicon der Verfalschungen u. Verun- reinigungen der Nahrungs- u. Genussmittel, der Colonialwaaren u. Manufacte. der Droguen, Chemikalien, u. Farbwaaren, gewerblichen u. landwirthschaften, v. Producte, Documente, u. Wortzeichen. Leipzig, 1886. KoNiG, J. — "Chemie der Menschlichen Nahrungs- u. Genussmittel." 2 Theil. Zweite Auflage. Berlin, 1886. 3^ Auflage, 1893. MoELLER, J. — " Mikroskopie der Nahrungs- u. Genussmittel aus dem Pflauzenreiche." 8vo. Berlin, 1886. GlEiARD, Ch., et DuPRE. A. — " Analyse des Matiiires alimeataires et Becherche de leurs Falsifications. Paris, 1894. PERIODICALS. Zeitschrift fiir Untersucliung von Lebensmitteln, &c., Eichstatt. Zeitschrift gegen Verfalschung der Lebensmittel, Leipzig. Vierteljahrsschrift der Chemie der Nahrungs- u. Genussmittel. Berlin, Analyst, London. YI.— THE PEESENT LAW IN ENGLAND RELATIVE TO ADULTERATION OF FOOD. The Sale of food and Drugs Act, 38 and 39 Vic, c. 63, 1878; ano Sale of Food and Drugs Act Amendjient, 1879, 42 and 43 Vic, c 30. § 39. The preamble of the "Sale of Food and Drugs Act" repeals the Acts in force relating to the adulteration of food. Section 2 defines the term food to include every article used for food or drink by man, other than drugs or water,* and the term " drug " to include every medicine for external or internal use. * This wide definition of food includes peppers, spices, flavouring-essences, none of which are foods, in the sense that they can be eaten alone ; but they are essential parts of compound foods, and as such come under the influence of the Act. On the other hand, such a substance as baking-powder (James V. Jones, 58 J. P. '230) is not included ; in the case quoteil, the baking- powder consisted of sodic bicarbonate 20 per cent., alum 40 per cent., and ground rice 40 per cent., and the vendor was prosecuted and convicted under section 3; but on appeal the conviction was quashed. Judge § 39.] THE PRESENT LAAV AS TO ADULTEHATIOX. 47 Section 3. No person shall mix, colour, stain, oi' powder, or order or permit any other person to mix, colour, stain, or powder, any article of food, with any ingredient or material so as tO' render the article injurious to health, with intent that the same may be sold in that state ; and no person shall sell any such article so mixed, coloured, stained, or powdered, under a penalty in each case not exceeding fifty pounds for the first offence ; Every offence after a conviction for the first offence shall be a misdemeanour, for which the person, on conviction, shall be imprisoned for a period not exceeding six months with hard labour. Section 4 is very similar to this, and relafes to drugs : " Xo- person shall, except for the purpose of compounding as herein- after descx-ibed, mix, colour, stain, or powder, etc., ifcc, any drug with any ingredient or material so as to affect injuriously the quality or potency of such drug, with intent that the same may be sold in that state, and no person shall sell any such drug so- mixed, coloured, stained, or powdei'ed, under the same penalty in each case respectively as in the preceding section, for a first and subsequent offence." The sections above quoted, foinnidable as they appear, possess. in reality no deterrent powers, but are perfectly harmless, since no prosecution is likely to succeed imder these sections, save when supported by very exceptional circumstances ; for the next section expressly pi-ovides that no conviction is to take place if the person accused " did not know of the ai-ticle of food Hawkins stating that, " the mere sale of an article, not in itself an article of food, even though it be sold in the knowledge of the vendor that it is the buyer's intention to mix with it the ingredients of which an article of food — e.g., bread — is to be composed, is no otfence under section 3, and it inakes no difference, in a legal point of view, that when sold, it is mixed with other ingredients not in themselves hurtful, some or one of which might in an unmixed state be used as articles or an article of food, if the injurious and the harmful articles are so inseparably mixed and in such quantities as that the mixture as a whole forms an injurious compound which no one would dream of usmg as a food." Alluding to the baking-powder in ques- tion, the learned judge said — " Wlio would venture to describe such a mixture as the food of man ? "With equal truth, might not powder com- posed of poison mixed with Hour be called food for man, because pure tiour is used? Possibly it may be said, that the injurious ingredients, when mixed with other materials of which an article of food is composed, become a part and parcel of siich article, but that is no argument against the vendor of such injurious ingredient, unless such injurious ingredient can be treated as an article of food at the time of sale. Tliat is the moment when the test of its character is to be applied, and if it is not then an article of food no offence is committed by the vendor of it, though the purchaser, or anyone who afterwards mixes it with an article of food intended for sale, would be guilty of au offence." 48 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 40. or drug sold by him being so mixed, coloured, stained, or powdered," and were able to show that he " could not with reasonable diligence have obtained that knowledge." § 40. The i-eal working sections of the Act are the following: — Section 6. No person shall sell, to the prejudice of the jntrchaser, -any article of food or any drug which is not of the nature, substance, and quality of the article demanded by such purchaser, under a penalty not exceeding twenty pounds, provided that an oftence shall not be deemed to be committed under this section in the following cases, that is to say : — 1. Where any matter or any ingredient, not injurious to health, has been added to the food or drug, because the same is required for the production or preparation thei'eof as an article of commerce, in a state fit for carriage or consumption, and not fraudulently to increase the bulk, weight, or measure of the food or drug, or conceal the quality thereof : 2. Where the drug is a proprietary medicine, or is the subject of a patent in force, and is supplied in the state required by the specification of the patent : 3. Where the food or drug is compounded as in this Act mentioned : 4. Where the food or drug is unavoidably mixed with some extraneous matter in the process of collection or preparation. In the amended Act, the second section, 41 and 42 Vic, c. 30, states that in any prosecution under the provisions of the principal Act for seUing to the prejudice of the purchaser any article of food or any drug, which is not of the nature, &c. , it shall be no defence to any such prosecution to allege that the purchaser, having bought only for analysis, was not prejudiced by such sale. Neither shall it be a good defence to prove that the article in question, though defective in nature or substance or quality, was not defective in all three respects. * The sixth section of the amended Act is to be read with the sixth section of the principal Act, for it states that " in determining whether an offence * Mr. Justice Mellor, in the case of Hoyle v. Hitchman, L. R. 4, Q. B. D. 2.S0, 43 J. P. 431, gave a judgment clearly expressing the meaning of pre- judice to the purchaser. " Tlie offence intended to be prevented by the Act was the fraudulent sale of articles adulterated by the admixture of foreign substances, which would necessarily be to the prejudice of the purchaser, and these words were inserted only to require that such adulteration should be shown to have been made; and further if the purchaser asks for a certain article, and gets an article which, by reason of some admixture of a foreign article, is not of the nature or quality of tlie article asked for, he is necessarily prejudiced. . . . The words 'to the prejudice of the purchaser' are necessary for the purpose of not interfering with the sale of an article of an inferior nature or quality to that demanded. Tlie prejudice contemplated was not confined to a pecuniary prejudice, for it would very much diminish the probability of bringing home otfences against the Act to those who were really guilty, and this was a sufficient argument against such a reading." § 41.] THE PEESENT LAW AS TO ADULTERATION. 49 has been committed under section six of the said Act, by selling to the prejudice of the purchaser spirits not adulterated otherwise than by the admixture of water, it shall be a good defence to prove that such admixture has not reduced the spirit more than twenty-five degrees under proof for brandy, whisky, or rum, or thirty-tive degrees under proof for gin."* Section 7 of the principal Act enacts that "No person shall sell any compoxmd article of food or compounded drug, which is not composed of ingredients in accordance with the demand of the purchaser ; penalty not exceeding £20." Section 8 provides " That no person shall be guilty of any such offence as aforesaid in respect of the sale of an article of food or a drug mixed with any matter or ingredient not injurious to health, and not intended fraudulently to increase its bulk, weight, or measure, or conceal its inferior quality, if at the time of delivering such article or drug he shall supply to the peison receiving the same a notice by a label distinctly and legibly written or printed on or with the article or drug, to the effect that the same is mixed." Section 9 enacts "That no person shall, with the intent that the same may be sold in its altered state without notice, abstract from an article of food any pai't of it so as to affect injuriously its quality, substance, or nature, and no person shall sell any article so altered without making disclosure of the alteration, under a penalty in each case not exceeding £20." § 41. One of the chief loopholes which offenders against the Act have diligently availed themselves of is the label section, Section 8. A label Avill often have a description of the article in large letters, such as COCOA, COFFEE, &c., and then in miserably small type a statement that the article is mixed. In the case of Liddiard v. Eeece (44 J. P. 233), a gi'ocerhadsoldhalf a pound of a mixture of coffee and chicory to an inspector ; the mix- ture was contained in a canister, and was duly weighed, and the full price of coffee was paid for it. After the sale had been com- pleted, the purchaser informed the appellant that he intended to have the article analysed. Thereupon, while the ])acket was still on the counter, the appellant called the purchaser's attention to the label, on which the purchaser noticed for the first time the words " This is sold as a mixture of chicory and coffee," printed in distinct and legible characters. The label was affixed in a con- spicuous position on the outside of the packet. The purchaser then said that he had asked for " coffee," and not for " chicory and coffee." The mixture was found by the analyst to consist of 60 * By section 14 of the Licensing Act, 1874, it is provided "That all convictions for selling adulterated drinks shall be entered in the proper register of licenses, and may be directed to be recorded in the license iu the same manner as if the offence had been committed against this Act. ' 5 50 foods: their comtosition and analysis. [§ 41. parts coffee and 40 parts chicory. On tbe hearing of the case before the magistrates, they convicted the vendor on tlie follow- ing grounds : — " The fact that the purchaser asked for coffee and was supplied with an article consisting of only 60 per cent, coffee and 40 per cent, chicory, without having his attention called to the label ; and without, in fact, seeing it until the purchase was completed, and also the fact that the price he paid for the said article was a usual and fair price for pure coffee, and much more than would have been given for coffee mixed with chicory to the above extent . . . and that, therefore, the appellant was not protected by the said eighth section." On appeal the case was decided in the Court of Queen's Bench, November 29, 1879, before Justices Lush and Manisty, who agreed with the magistrates, and the conviction was affirmed. There is also a similar decision in the case of Horder v. Meddings, 44 J. P. 234.* In Jones v. Jones, 58 J. P. 132, it was decided that it was not necessary to call the buyer's attention to the label, provided the label is distinct and that there is no fraudulent concealment as to the low quality of the article sold. The case was that the vendor sold a mixed cocoa, the packet being legibly labelled as a mixture, and it was held that the seller was pi-otected by such label Any verbal declaration is no protection unless it is uttered before the sale is completed. The sale, again, is evidently not completed until the goods are delivered into the purchaser's hand, and the vendor has received the money. Should a person buy any substance in a shop, and (after having * This case probably overthrows the case reported in the T'/mes, June S, 1S79, Gibson v. Leaper, a prosecution undertaken under tlie old Act, 35 and 36 Vic, c. 74, sections 2 and 3. On conviction the vendor appealed. The case was that of a Spalding grocer, who sold a packet of " Epps's Cocoa" with- out making any verbal statement of its contents. The packet was labelled with the words " prepared cocoa — for ingredients, see the other side," and on the other side was a notification to the effect that it was necessary in order to make the oil in the cocoa soluble and easy of digestion, to combine with it arrowroot and sugar. The court quashed the conviction, holding that assum- ing the cocoa to be adulterated, it had not been sold as unadulterated. In the case of Pope r. Turle (43, Law Journal), ^lay 2S, 1S74, the Justices of Bed- ford dismissed a summons for selling adulterated mustard, and the purchaser appealed. It was stated in the case that at the time the respondent delivered the mustard to the appellant he said : " I do not sell you this as pure mustard." The mustard was found to be the common mixtui-e of flour and mustard. Lord Coleridge, Mr Justice Brett, and Mr Justice Grove, were undivided in their opinion that the seller was entitled to their judgment on the ground of his having declared to the purchaser that the mustard was mixed with some other ingredient, and that, even had he not done so, he could not come within the section to incur the penalty, because if the admixture was such as to make it an adulterated article, still he had not sold it as an unadulterated article. § 42.] THE PRESENT LAW AS TO ADULTE RATIOX. 51 tendered his money, and the same has been accepted) proceed to state that the article is required for analysis, and the vendor then attempt to return the money : if the purchaser does not accept the money, the sale is evidently comj)lete. On the other hand — an inspector went into a druggist's shop and asked for quinine wine. The chemist served him with the wine, wrapped it up, and laid it on the counter. The inspector then produced his bottles, and declaring the nature of his errand, was about to divide the wine into three parts, when the druggist seized the bottle and refused to sell the wine, which, a moment before, by his actions he seemed ready to do. In this case, the sale w^as not complete. But now, let us suppose that the inspector had been a little quicker than the chemist, and seized the sample, and, notwithstanding the expi-essed refusal of the druggist to sell, the inspector had cast his money on the counter — Would the drug have been sold 1 This question is somewhat difficult to answer, but I think that it would have been a sale, and, if adulteration had been proved, the vendor would probably not have escaped through adopting the defence that there had been "no sale."* In the case of a grocer who sold adulterated coffee, the vendor had received the money, and had laid the packetand also the change on the counter, but on hearing the errand of the purchaser he laid his hand on the change and the packet, declaring that the sale was not complete, as he had not given the change, and also that he did not sell the goods as unadulterated. But the magis- trates very properly did not admit the defence. § 42. There is an important question as to how far a vendor can be protected by having a board in or over his shop or place of business, giving notice to the effect that all the goods are adulterated. The English law is made for those who cannot read as well as for those who can : and presuming a purchaser to be uneducated, the notice gives him no information. Again, it is certain that a very large number of purchasers, even should the notice be in a conspicuous place, fail to observe it. In most cases in actual practice such notices are distinct evasions of the Act, being inconspicuous, and in dark corners. A seller of milk had a van on which a notice was placed, " Country skimmed milk, sold as adulterated milk." The man with his can went on foot from door to door, the van being in the road. It is evident that, in such a case, very few of the customers could have seen the label. An inspector wdio bought * In any case, the druggist might have been prosecuted under sect. 17 of the principal Act, for refusing to sell. 52 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 43. a sample of the milk did not see it, and the magistrate convicted the defendant.* The important appeal case of Sandys v. Small,f decided before the Court of Queen's Bench, June 25, 1878, bears upon this and lays down the law. A publican put up a notice in liis house : "AH spirits sold here are mixed." The inspector of weights sent a messenger to buy some whisky, which was given without anything more being said on either side ; but the purchaser admitted that before he bought the whisky he saw the notice, " All spirits sold here are mixed, 38 and 39 Vict., c. 63, sec. 8 and 9," although at the very moment of buying the whisky he did not see it. It was proved that a similar notice ■was posted at the bar window in full view of persons purchasing. Chief-Justice Cockbui-n said : — " If the seller chooses to sell an article with a certain admixture, the onus lies on him to pi'ove that the purchaser knew what he was purchasing. With respect to the alteration of the article, the Act has provided him with the means of protecting himself against such a presumption, and says that if he attaches to tlie article a notice of the adultera- tion which has been made in its quality, then he shall be pro- tected against any charge of an ofi'ence against the Act. If he does not resort to this protection, then the presumption of law attaches, and is unrebutted. If he can show that he brought home by otlier ways to the knowledge of the customer, that tlie quality of the article was altered by admixture, then he does not commit tlie offence, because both parties knew it, and the seller does not sell an article to the prejudice of the purchaser, and the parties are pei*fectly free to contract on that footing. In that view the seller, if he has stuck up a notice, would not commit an offence though he might not have affixed a label to the bottle, because he did not sell 'to the prejudice of the purchaser.' . . . It was sufficiently manifest that the man who was sent to buy the whisky knew of the notice stuck up, and hence it Avas clear that the defendant committed no ofience." From this judgment it is sufficiently evident that where tlie general label or notice has been clearly seen and understood before making the purchase, then no offence is committed. The decision of Liddiard v. Reece does not cover exactly the same ground as the case just quoted, but both, I think, support the view here put forward — viz., that the defendant is bound to prove that the purchaser had a clear knowledge of the quality of the goods before purchasing. § 43. Section 10 provides for the appointment of public analysts in England, Scotland, and Ireland, by various local * Analyst, 1880, p. 225. f L. R. 32 B. D. 449 ; 42 J. P. 550. § 43.] THE PRESENT LAW AS TO ADULTERATION. 53 bodies, such as, in England, the Commissioners of Sewers for the City of London, the District Councils of the Metropolis, the County Councils of Counties, and the Town Councils of Boroughs with a separate police establishment ; in Scotland, the Commissioners of Supply, or the Commissioners of Boards of Police, or, where there are no such Commissioners, the Town Councils of Burghs ; and in Ireland, the Gx'and Juries of the Counties and the Town Councils of the Boi'oughs. These appointments must be confirmed by a central authority, which, in England, is the Local Government Board ; in Scot- land, one of Her Majesty's Secretaries of State ; and in Ireland, the Irish Local Government Board. The appointment is, in the first instance, i^ermissive, but the superintending or central authority may compel the appointment, and the tilling of any vacancy appears to be compulsory. The qualifications of the analyst are, to a certain extent, defined by the Act, for it directs that there shall be appointed " one or more persons possessing competent knowledge, skill, and experience." It has been thought that the person appointed must have had a medical education ; but although this may be desirable, and extremely useful, yet it is certain that with regard to the carrying out of the Act itself, the best qualifications are those of a chemical and scientific nature. A board selecting an analyst for the first time should insist more especially on chemical experience, as evidenced by original work, and the having passed such an examination as that of the Institute of Chemistry. It is a most serious thing for the traders of a town or county to be at the mercy of incompetence or inex- perience, and many of the appointments which were at first made under the Act were so notoriously vinsuitable, that a great deal of undeserved odium was thrown upon the whole body of analysts. Lately, however, the " survival of the fittest " process has been going forward, with the result of a great improvement, and one likely to be continuous, more especially as the Local Government Board, acting under skilled advice, is now very cautious in confirming appointments, and insists upon proper qualifications. The eleventh section distinctly lays down the principle of combi- nation, enacting that the town council of any borough may unite with that of any neighbouring borough in appointing an analyst jointly ; or the analyst for the county in which the borough is situated may act upon arrangement as analyst for their borough. Those who ai"e practically acquainted with the subject know, that it is only in the largest and most populous places in Eng- land that any kind of living can be made out of a public analy- 54 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 44. tical appointment. Hence, it follows that an analyst for a small place must either haA'e private means or that his chief occupa- tion must be of a more remunerative nature ; it is, therefore, highly desirable that the analysis of foods and drugs should be in a few hands only, and that an analyst should hold many appointments of the same nature. In this way, and in this way only, will it be possible to have properly fitted laboratories, supplied with all the expensive appliances of modern research, and in this way only will it be possible to improve the processes of analysis. It is also a fact, from the very few cases in which an experienced analyst has to attend as witness, that there would be no inconvenience, were all the northern counties to have their samples analysed at Sheffield, Manchester, or Yoi'k ; the western and south-western counties at Bristol ; and the rest of England at the London laboratories. Probably also the whole of the Scotch samples could be dealt Avith in Glasgow and Edinburgh, and the Irish, in like manner, in two of the chief cities. § 44. Section 12 of the principal Act provides for tlie purchase of samples by any purchaser for analysis by the public analyst for the district in which the purchase is made, on payment to such analyst of a sum not exceeding ten shillings and sixpence; or if there is no analyst appointed for the district, to the analyst of another place. In this latter case the fee appears to be a matter of private arrangement, for the a\ ords of the Act are — " such sum as may be agreed ujjon between such person and the analyst." In either case, the analyst must give a certificate of his results to the purchaser. Whether the purchaser is bound to purchase the article in the manner directed in section 14, is still an undecided point. It is evident that, for legal purposes, the official analyst must be employed, and that under the Act no prosecution can be undertaken except on his certificate. Thus, at the Manchester Police Court, the Milk-Dealers' Protection Society attempted to prosecute on the certificate of a private analyst, but on this technical ground alone the magistrate dismissed the case.* There is no authority given by the Act for a Public Analyst himself to appoint a deputy. At Bristol the analyst sufi"ering from illness deputed an analyst in a neighbouring district to do his work. The dej)uty certified to the adulteration of a sample of milk. The defendant took advantage of this omission in the Act and successfully disputed the power of an analyst to appoint, even under the circumstance of illness, a deputy to do his statutory duty. Whether this case should have been otherwise decided on appeal is not clear, for the next section gives power to a pur- * Analyst, 1879, vol. iv., p. 74. § 45.] THE PRESENT LAW AS TO ADULTERATION. 55 ■chaser to take a sample to " the analyst of another place " should there be no analyst "acting for such place"; and it might be reasonably argued that an analyst disabled by illness is for the time "not acting for such place." § 45. The thirteenth section of the old Act and the third section of the amended Act should bo read together: — " Any medical officer of health, inspector of nuisances, or inspector of weights and measures, or any inspector of a market, or any police-constable under the direction and at the cost of the local authority appointing such officer, inspector, or constable, or chai'ged with the execution of this Act, may procure any sample •of food or drugs, and if he suspect the same to have been sold to him contrary to any provision of this Act, shall submit the same to be analysed by the analyst of the district or place for which he acts; or if there be no such analyst then acting for such place, to the analyst of another place, and such analyst shall, upon receiving payment as is pi-ovided in the last section, with all convenient speed, analyse the same and give a certificate to such officer, wherein he shall specify the result of the analysis." By section 3 of the amended Act, the same individuals " maj procure at the place of delivery any sample of any milk in course of delivery to the purchaser or consignee, in pursuance of any contract for the sale to such purchaser or consignee of such milk." Section 4 of the same Act provides a penalty for refusal to submit samples of milk to be taken, of a sum not exceeding £10. Section 17 of the principal Act also provides a penalty not exceeding £10, for refusal to sell to the persons appointed to carry oiit the Act any " article of food or any drug exposed for sale, or on sale by retail on any premises, or in any shop or stores. The purchaser shall tender the price for the quantity which he shall require for the purpose of analysis, not being more than shall be reasonably'- requisite." Any street or place of public resort is held to come within the meaning of this section. It is pei'fectly clear from the sections quoted, that if a sample be taken of milk in transit, that sample must be taken at the place of delivery. If, for example, a milkman is driving his cart through Oxford Street, it would not be legal for an inspector to stop the cart and require a sample of the milk. The sample must be taken at the place where the milk is delivered. This may be a house, or it may be a railwa3^-station, or it may be a public booth where the milk is sold at so much a glass. From the case of Eouch v. Hall * heard before the Court of Queen's Bench, it is evident that in procuring samj)les * L. R. 62 B. D. 17; 50 L. J. 6; 45 J. P. 220. 5G FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 46. at the place of delivery, the inspector need not divide the sample. The case was briefly as follows : — The inspector was at Euston Station, and saw a can of milk taken from the van. He accordingly demanded and received a sample, and treating the porter as the agent of the respondent, divided the milk into three parts, and gave one of the parts to the porter, telling the latter that he intended to have the milk analysed. He took no steps to inform either the respondent or the consignee of his intention, but on finding the milk adulterated with water laid his informa- tion. The case was dismissed by the magistrate, and the inspector appealed. Mr. Justice Field said that the appeal must be allowed. The Court was clearly of opinion that the railway porter was not the agent of the respondent within the provisions of the fourteenth section, nor was he bound to accept a third of the sample of the milk, although he would have been liable to a penalty had he refused to supply a sample. The object of the Act was to secure to the public a supply of pure, unadulterated milk, and for that purpose it w-as liable to seizure at the time of its being sold by the seller or his agents, provided that a third of the same sample should be tendered to him, so that he might be enabled to have an independent analysis to show whether it was adulterated or not. But as milk had to be supplied from the country, and it was found that a hardship was often inflicted on the London seller, to whom adulterated milk was supplied by farmers, it was enacted by the Amendment Act of 1879, 42 and 43 Vict., c. 30, that the inspector should have the power of seizing the milk at the place of delivery to the consignee. In this case, although the delivery had not been completed, and although the railway porter could not be held to be the agent of the consignor, the Court was of opinion, that by the Amendment Act the legislature did not intend to extend to the consignor that privilege which was aff"orded under the previous Act to the seller, namely, that of giving him a third of the sample to enable him to obtain an independent analysis. The case was then remitted to the magistrate in the usual form. § 46. Section 14 fully details the method to be pursued by any purchaser under the Sale of Food and Drugs Act. The person purchasing any article with the intention of sub- mitting the same to be analysed, shall after the purchase shall have been completed, forthwith notify to the seller or his agent selling the article, his intention to have the same analysed by the public analyst, and shall ofl'er to divide the article into three parts to be then and thei'e separated, and each part to be marked and sealed up or fastened up, as its nature will permit, and shall if required to do so, proceed accordingly, and shall deliver one of § 46.] THE PRESENT LAW AS TO ADULTERATION. 57 the parts to the seller or his agent. He shall afterwards retain one of the said parts for future comparison, and submit the third part, if he deem it right to have the article analysed, to the analyst. Section 15. If the seller or his agent do not accept the offer of the purchaser to divide the article purchased in his presence, the analyst receiving the same article for analysis shall divide the same into two parts, and shall seal or fasten up one of those parts, and shall cause it to be delivered, either upon receipt of the sample, or when he supplies his certificate to the purchaser, who shall retain the same for pi-oduction, in case proceedings shall afterwards be taken in the matter. In the case of Horder v. Scott,* heard in the Queen's Bench division before Justices Lush and Field, on the 4th of May, 1880, it was made clear that an inspector could appoint a deputy. The case was an appeal from a decision of justices in the county of Stafford. An inspector vmder the Act had deputed his assistant to purchase a sample of coffee, which was duly divided in conformity with the Act, and the analyst certified to its adulteration with chicory. The magistrates, however, considered! that as the proceedings were initiated by the inspector in his official capacity, he having laid the information, and having regard to sections 13, 14, and 17 of the Act, should personally have purchased the article, and the case was dismissed. This, Mr. Justice Field said, was entirely wrong — " It did not signify whether the inspector purchased by his own hand or by his agent. Then the magistrates had decided, secondly, that Samuel Toy, being the purchaser, should have submitted the article to the county analyst ; there again he thought the magistrates were wrong. ... If the thing wei'e properly analysed, it does not signify through whose hands the article was bought." On the purchase of an article it is evidently essential to say, not only that it is the purchaser's intention to have it analysed, but " analysed by the public analyst," care being taken to use the exact woi'ds of the Act. This objection has been several times raised with effect. When a deputy purchases samples, it would be a mistake for the inspector to appear and seal the samples. This had better be left to the purchaser, who can then immediately, or at any subsequent period, hand the samples to the inspector, by whom they should be delivered to the analyst. It is obvious that legal proof will be required as to the proper keeping and delivery of the samples. It has been argued that the division of the sample into three parts means three equal parts ; but there is no direction in the * L. R. 52 B. D. 552 ; 49 L. J. 7S ; 44 J. P. 520, 795. 58 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 47. Act as to an equal division. At the same time, sliould the purchaser leave with the seller, or keep himself an insuffi- cient quantity for any further analysis, there would be an infringement of the spirit of the Act ; for the purpose of the division evidently is to provide against any mistake or wrong interpretation of facts on the part of the analyst. Should another analysis be required, it would not be right that the seller should be put at a disadvantage by any marked or great inequality in the division of the parts ; hence it will be prudent for purchasers to divide the substance into three parts as nearly equal as may be, but it is imnecessary to use for this purpose balances or measures. On the seller or his agent not accepting tlie offer of the pur- chaser to divide the sample into three parts, it becomes the duty of the analyst to divide it into two parts. Tliere is no direct stipulation as to when this is to be done, for the analyst is permitted to keej:), if he cliooses, the whole, until the termination of the analysis ; but it is evidently the course most free from objection to divide it into two approximately equal parts immediately on receipt of the sample, to seal it in the presence of the purchaser, and deliver one of the parts to the purchaser. § 47. Section 16 permits articles to be sent by post after being duly registered, and the Postmaster-General has made the follow- ing regulations witli i-egard to the transmission of samples : — ^'Each packet must be addressed according to the official designa- tion of the analyst, as 'public analyst,' or otherwise; the nature of its contents must be stated on the front of the packet. Any postmaster, at whose office a packet for a public analyst shall be tendered for registration, may refuse to accept it for this purpose, unless it be packed in so secure a manner as to render it at least unlikely that its contents will escape, and injure the correspondence. Liquids for analysis shall be contained in stout bottles or bladders, which shall be enclosed in strong wooden boxes with rounded edges — the boxes being covered by stout wrappers of paper or cloth, and no such jiacket shall exceed eight inches in length, four in width, or three inches in depth. No packet whatever addressed to a public analyst shall exceed the dimensions of eighteen inches in length, nine inches in width, or six inches in depth. The postage and registration-fee on each packet must be prepaid." As analyst for a distant county, I have had made a large number of small wooden boxes for the purpose of transmitting samples, and these I have supplied to the inspectors. In this •way I have received samples of milk, cream, butter, wines, sjpirits, and other matters through the post for several years, and § 48.] THE PRESENT LAW AS TO ADULTEUATlOlSr. 59 no difficulty has been experienced. Bulky matters, sucli as beer, loaves of bread, and other substances, -which cannot be sent by ordinary post, can, of course, be forwarded by the parcel post; the introduction of which has been a great boon to all those in analytical practice. § 48. Section 18 states that the cei'tificate shall be in the form set forth in the schedule, or to the like effect. These last few words are important, for the analyst thereby is not absolutely confined to the certificate in the schedule. Notwithstanding this, it is safer to adhere strictly to the exact form of certificate, and not to attempt to modify it in any way. In certifying, the more definite the certificate is the better. An analyst having given a certificate as follows : — " A samj)le of coff"ee was adulterated with 20 per cent, of vegetable matter, which I believe to be chicory," the magistrate dismissed the case, on the simple ground of " the loose wording of the certificate." Probably the magistrate was wrong, for if the Avords meant anything at all, they meant that the cofi"ee was adulterated with some vegetable ingredient that, whatever it was, was not coS'ee. Nor do I see that it is essential for the analyst to know the exact nature of a substance added, so long as he is perfectly clear that the substance is foreign to the article, and not of the nature that the purchaser demanded. It is obvious that coffee may be adulterated by some foreign root which no analyst has ever seen or heard of; and it would surely be a certificate to be accepted and adjudicated upon if the analyst (under these circumstances) were to certify, "This coffee is adulterated with 20 per cent, of vegetable matter which is not coffee, but the exact nature of which is unknown to me." Again, an analyst certified — '■ Practically, all chicory," and the magistrates dismissed the case on the ground of "the loose word- ing of the certificate." Here it is probable that the magistrates were right, for such a certificate is neither in the form nor to the efiect of the certificate appended in the schedule to the Act, which plainly implies that where there is adulteration, the analyst shall state the pei'centages of parts. It is true that the case might have been adjourned for the attendance of the analyst, or the certificate might have been amended, but nothing in the Act contemplates or provides for any inaccuracy or carelessness in drawing out the certificate. In the case of any article liable to decomposition, the analyst must certify specially as to whether " any change has tahen place in the constitution of the article that loould interfere vjith the analysis." Milk and butter are specifically mentioned, but the rule would evidently apply to all foods preserved in tins, provided the tin has been opened. It might also be argued that many other 60 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 49. substances (such as wine or beer) are liable to decomposition ; hence, it will be better for the analyst to give this matter rather a wide interpi-etation, and insert in his certificate the necessary words, if called upon to certify in reference to any substance that, under any conditions, is liable to decompose. The exact words must be used, for in an appeal heard at the Middlesex Sessions, October, 1880, Peart v. Edwards,* the analyst certified that the milk was fresh when delivered to him, bvit omitted to specify whether " any change had taken place in the constitu- tion of the article, so as to interfere with the analysis ; " and on this ground the assistant-judge quashed the conviction. Section 19 provides for the regular quarterly reports of the analyst, copies of which are to be transmitted to the Local Govern- ment Board. If, as in many cases, no work at all has been done under the Act, it is evidently the duty of the analyst to send a " nil " report. Section 20 provides for the institution of proceedings. The Act says — "The person causing analysis may take proceedings." He, therefore, need not be the actual purchaser ; and it is usual for an inspector to take the summons out on behalf of the public body for which he acts. In all prosecutions imder the Act, and notwithstanding the section just quoted, the summons must be served within a reason- able time, and in the case of a perishable article, e.g., milk, not exceeding twenty-eight days from the time of the purchase, &c. The summons must state the particulars of the oflience or oflences, and also the name of the prosecutor ; and it must not be made returnable in less than seven days from the day it is served upon the person summoned. Section 21 of the principal Act provides that the certificate of the analyst shall serve as evidence ; therefore, unless speci- ally required, he need not attend. If, however, the defendant require the analyst to be called as a witness, he will then be obliged to appear. This request for the analyst to attend may be by notice from either the solicitor or the defendant himself, or it may be by request in court at the first hearing of the case, in which instance, the case will probably have to be adjourned. Such notice should certainly be given in writing to the analyst, but still it is not advisable to ignore a verbal request. § 49. Section 22 provides for a part of the sample, or sam})]es, to be analysed at Somerset House, in case of any dispute as to the coi-rectness of the analysis. The sample must be sent by order of the justices (or a stipendiary magistrate), and would not be received if sent by either party direct. The results of * 44 J. P. 699, 763. § 49.] THE PRESENT LAW AS TO ADULTERATION. 61 tlie official analysis should be considered, as the written opinion of a i-eferee not final but yet entitled to great weight. Defen- dants, notwithstanding this clause, are very fond of employing private analysts for the defence : certainly a most unwise pro- ceeding, for if the analysis is disputed, power is given under the Act to refer the matter to a laboratory, which, from the very nature of its constitution, will be perfectly impartial, and the certificate of which will be admitted as evidence. Section 23 provides for an appeal to Quarter Sessions. Section 25 gives the opportxmity to the defendant to prove by wi'itten warranty, " that he had no reason to believe at the time when he sold it that the article was otherwise than of the nature, quality, &c., demanded; that he sold it in the same state as when he purchased it." On pi'oof of this, the defendant may be dis- charged from the prosecution, but he will have to pay costs, unless he has given notice to the prosecutor that he will adopt this line of defence. In the case of Rook v. Hopley,* it was decided that an invoice containing a description of an article sold to a retail dealer is not such a written warranty as is required by Section 25 ; and a retail dealer who sells an adulterated article in the same state as he pui'chased it will not, by virtue of such a document, be entitled to be discharged on being summoned before a magis- trate. Section 26 provides for the payment of penalties i-ecovered, to the authority, for the purpose of defraying the expenses of the Act. Section 27 has stringent clauses relative to persons convicted of forging warranties, wilfully applying a certificate or warranty of an article of food or drug, to any other article of food or drug, the giving of a false warranty, and wilfully giving a label falsely desci'ibing the article sold. This latter clause of the section — viz., "Every person who shall wilfully give a label with any article sold by him, which shall falsely describe the article sold, shall be guilty of an offence under this Act," &c. — would apply to a great many cases of adultera- tion in which the article is wrongly described by label; but it is evident that guilty knowledge must be proved, for the word "wilfully" presupposes guilty knowledge. In most cases, unless the actual manufacturer were summoned, ignorance would be pleaded. Section 28 provides that nothing in the Act shall aflfect the power * L. R. 34 D. 209 ; 42 J. P. 551. (Footnote page 50.) An experienced legal friend does not agree with this, but considers that whether a label protects or not depends on the amount of adulteration. 62 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 50. of proceeding by indictment, or take away any other remedy against any offender under the Act, or in any way interfere with contracts and bargains between individuals, and the rights and remedies belonging thereto, provided that in any action brought by any person for a bi-each of contract on the sale of any article of food or any drug, such person may recover alone, or in addition to any other damages recoverable by him, the amount of any penalty in which he may have been convicted under this Act, together with the costs paid by him upon such conviction, and those incurred by him in and about his defence thereto, if he prove that the article or drag, the subject of such conviction, was sold to him as and for an article or drag of the same nature, sub- stance, and quality as that which was demanded of him, and that he purchased it not knowing it to be otherwise, and afterwards sold it in the same state in which he purchased it — the defendant, in such action, being nevertheless at liberty to prove that the conviction was wrongful, or that the amount of costs awarded or claimed was unreasonable. The 30th section of the Act provides for the examination of tea on importation. The eflect of this examination has been so good that adul- terated tea, in compai-ison with the period before the Act, has decreased in a very marked degree. VIL— THE DUTY OF THE INSPECTOR, OR PURCHASER UNDER THE ACT. § 50. It will be the duty of the inspectors appointed under the Act by the local authority employing them, to take and submit samples from time to time to the public analyst, and it will greatly depend on their intelligence and activity whether the Act will be carried out ])roperly or not. An active inspector, if he is not known when ho commences the work, will soon become so, and it will be necessary to employ, as a rule, deputies. The deputies, it is hardly necessary to state, should not be childx'en, but intelligent adults of either sex, and they should be carefully instructed in the "purchase clauses" of the Act, and taught how to seal and properly divide the samples purchased. It will be necessary for the official purchaser to carry with him all materials for properly labelling and sealing samples. A convenient bag with bottles, jars, wrapping paper, wax tapers, matches, sealing-wax, and an official seal, will therefore be essential. § 50.] THE DUTY OF THE INSPECTOR. 63 The sample should not be divided, noi' any declaration made until the sale is absolutely complete and the sample in pos- session of the inspector ; when that is the case, the exact words of the Act must be used, and he must say, " I have bought this for the purpose of having it analysed by the jyiiblic analyst" and then he must offer to divide it into three parts, which he will at once proceed to do, unless the seller decline to take advantage of the offer; in that case the purchaser will take the whole to the analyst. The purchaser must carefully note any declaration which the seller may make with regard to the article, and especially whether such declaration is made before or after the completion of sale. The division of the sample must be as equal as possible, and the parts must be very carefully sealed. In sealing bottles, the cork should be driven in flush with the surface of the neck, and the seal not only placed on the top of the cork, but carried round on to the neck itself, so as to render it impossible for a knife to be inserted under the wax and the cork removed without breaking the seal. A label identical in wording and number must be affixed to each division of the same sample. Tn the case of butter and substances which cannot be put into corked bottles, the best method is to wrap the substance, or jar containing the substance, in paper, and put several seals on the paper in such a way that it is impossible for the packet to be tampered with.* The inspector should always carry a copy of the Act with him, and in case of a refusal to sell, he should then present his card, or declare that he is an inspector didy appointed to carry out the Act, and call the attention of the seller to Section 17, and tender the price of the article sold.t If the seller still refuses to sell, then the pui'chaser evidently has a case under the Act, and should proceed accordingly. The official purchaser should not select to the exclusion of others the pooi'est shops, but take samples as equally as possible. The purchase of samples need not be effected in an officious mannei", nor is it just, for example, to enter a shop when full of people, and with ostentation buy and divide the sample before the customers, for an injury may thus be done to an honest tradesman ; the people in the shop might naturally think, in such a case, that the tradesman's goods were " things suspect." There are indeed always two ways of doing a thing, and a little politeness and civility will in no way interfere with the execution of duty, or the carrying out of the Act. * Or better still to use strong paper or canvas bags. f The Inspector must actually tender the money, not merely show a purse or offer to pay. 64 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 50, The official purchaser will probably be abused occasionally in no measured terms, but he must endeavour to keep his temper, and in no Avay retort. The sample retained by the purchaser must be locked up in a drawer or place to which no one else could have access without the key. Properly speaking, articles like milk apt to decompose should be kept in an ice safe. Inspectors should from time to time consult the analyst as to what samples would be advisable to take for analysis. There are many substances — e.g., white sugai'— which are so seldom adulterated that it is scarcely worth while obtaining samples of such, unless there has been some information laid relative to their quality. In taking samples of milk in the street, as before stated (p. 55), it is of no use for the inspector to stop the milk-seller while actually carrying his cans from door to door, but he must buy it at a place of delivery ; for example, he could not take a sample from an itinerant milk-seller legally while the milk-seller was going from one door to another, but directly the milk-seller stops at any door, he may then demand a sample and tender the money -for it, because then the milk is being delivei-ed. He may also go to a railway station, and take samples of milk from the cans them- selves; in the latter case, it does not appear necessaiy to divide the sample into three parts, but the analyst will be obliged to divide it into two, and give the inspector one (see p. 57). P^RT II. INTRODUCTORY. PART ll.-INTRODUCTORY. A DESCRIPTION OF A FEW SPECIAL FORMS OF APPARATUS USEFUL IN FOOD-ANALYSIS. § 5L As stated in tlie first edition of this work, it is no part of the author's plan to describe the elementary apparatus to be found in every text-book, and to be seen in every laboratoiy. Notwithstanding, it will be convenient here to give a brief notice of some special forms of apparatus useful in food-analysis. APPAEATUS FOR THE TREATMENT OF SUBSTANCES BY VARIOUS SOLVENTS. It is a matter of some moment to economise alcoholic and ethereal solvents, and it is always advantageous to keep a laboratory as free as possible from vapours and odours. Where a solid has to be exhausted by ether or petroleum, one can scarcely imagine anything more convenient than the ap- paratus invented by Soxhlet, and proposed by him for the purpose of treating milk solids with ether, but in point of fact widely applicable. It consists of a glass tube (fig. 1), the size of which is per- fectly under control, and may be made very large or very small, according to individual require- ments. For the purpose of milk analysis a capacity of 100 cc. is ample. The tube is quite closed at the bottom, A ; the volatile vapours ascend through the tube D, and are condensed in an upright condenser attached to A; the liquid, therefore, falls drop by drop on to the substance at A. When the condensed liquid reaches X, the syphon BB acts, and the whole of the liquid runs into the flask. The apparatus works quite automatically, and scarcely any ether is lost, how- ever long the operation may last. Clausnizer has modified this apparatus for small quantities of substances (tig. 2). D is a tube drawn out pipette-like. FOODS : THEIR COMPOSITION AND ANALYSIS. [§51. J is an Inner tube made so as to slip easily into the outer one, and pulled out in the blowpipe flame into a long, almost capillary stem, •which is then bent up into a syphon, and terminates in the flask, being made a little longer than the drawn-out portion of D. The mode of action is precisely similar to that just described. The volatile vapour escapes between the two tubes, until it reaches the upright condenser; it is then condensed, and falls in drops on the substance in the inner tube ; and when the bend of the syphon is reached, the little tube is at once emptied of the satui-ated solvent, and the process commences again. Another method is as follows (see fig. 3): — Take a flask with a wide neck, fit a small short test-tube, T,into a cork that will go tightly down into the neck of the flask ; cut two or tlii'ee notches in the cork, as shown in the figui-e. The ether or other solvent tl ^JSMj^ continually drops on to the substance J wiwlii ^^ ^^^^ tube, and when the tube is full, runs over into the flask, and thus the substance is at length exhausted. Similarly, any little apparatus may be suspended fx-om the cork by means of a wire ; or, lastly, a little tube may be supported from the bottom by means of a platinum wire-support fused into the flask. In the Extraction of Liquids by ether, petroleum, &c., the author has found the following ap- paratus absolutely indispensable, when it is necessary to make, by this means, any quantitative estimations (fig. 4): A is a tube of any dimensions most convenient to the analyst. Ordinary burette size will perhaps be the most suitable for routine work; the tube is furnished with a stopcock and bent tube B, the tube A having a very small but not quite capillary bore. The lower end is attached to a length of pressure tubing, and is connected with a small reservoir of mercury, moving up and down by means of a pulley. To use the apparatus : Fill the tube with mercury by opening the clamp at H, and the stopcock at B, and raising the reservoir until the mercury, if allowed, would flow out of the beak. Now, the beak is dipped into the liquid to be extracted with the solvent, and by lowering the reservoir, a strong vacuum is created, which draws the liquid into the tube ; in the same way the ether is made to follow. Should the liquid be so thick that it is not § 51.] APPARATUS FOR FOOD-ANALYSIS. possible to get it in by means of suction, the lower end of the tube is disconnected, and the syrupy mass worked in through the wide end. When the ether has been sucked into the apparatus, it is emptied of mercury by lowering the reservoir, and then firmly clamped at H, and the stop- cock also closed. The tube may now be shaken, and then allowed to stand for the liquids to separate. When there is a good line of demarcation, by raising the reservoir after open- ing the clamp and stopcock, the whole of the light solvent can be run out of the tube into a flask or beaker, and recovered by distillation. For heavy solvents (such as chloro- form), which sink to the bottom, a simple burette with a fine exit tube is preferable ; but for petro- leum ether, ordinary ether, «tc., the apparatus figured is extremely useful. When it is necessary to treat substances in open dishes with volatile solvents, the author uses the following apparatus (see fig. 5), which, since the first description of it in the Chemical Society s Journal* is to be found in most laboratories. The principle of the arrangement is simply this, that it converts an^ ordinary dish into a closed vessel, Y\ Leipzic, 1877. so FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 58. powers, is excellent. It merely consists (see fig. 1 5) of a conical tube, turned out of boxwood, and fastened on to the microscope tube bj means of a screw, h. Combined with this funnel tube is a wooden tray, in which the frame, e, d, f, g, easily slides backwards and forwards. Thin panes of glass are let in the cassette into this frame. Whilst the image is being adjusted, the thin glass, e, d, stands over the tube, and the prepared plate is put under the little cover at g, f. If the picture is well defined, the frame, e, d, g, f, is pushed into the tray, so that the part, g, f, ■can stand over the microscope tube, and by a simple ai-rangement the photographic plate can be exposed. Direct sunshine will, in most cases, be necessary, and the rays should be transmitted through a cell containing the ammonia-sulphate of copper. If, however, it is desired to photograph with high powers, the plan recommended and employed by Dr. Woodward, of the Army Medical Museum, Washington, is perhaps the best. The camera box and table are both dispensed with, and the operating room itself is converted into a camera. A room is selected having a southern aspect; the window is provided with shutters on the inside to exclude light, sufiicient being admitted through one or two yellow panes to enable the operator to move about freely; a small yellow pane is also let into one of the window-shutters to enable the operator to watch the face of the sky. The microscope is placed horizontally, and a heliostat outside the window throws the direct rays of the sun on to the mirror. The frame of the plate-holder runs on an iron track, ten feet long, and laid on the fioor at right angles to the plane of the window. There are most ingenious arrangements for working, although at a distance, the fine adjustment of the microscope; the sun's rays pass through a solution of the ammonia-sulphate of coppei'.* The fixing of the picture iipon the plates, the method of printing from the nega- tives, &c., are all extremely simple operations (especially to those accustomed to chemical manipulations), and are well described in standard treatises on photography. § 58. Colour. — It will often be necessary to ascertain the exact colouring-matter used to make articles of food attractive, more especially confectionery, jellies, pickles, &c. The question will generally resolve itself into deciding as to whether the colour is harmless or poisonous, and, if the latter, whether the poison is in sufficient quantity to injure the consumer's health. The poison- ous coloiiring-matters are those containing lead, copper, arsenic, ■chromium, and zinc, all of mineral origin; together with a few * The whole arrangement is figured and described in Dr. Beale's work on the microscope. § 58.] COLOUR. 91 injurious organic colouring substances, such as gamboge and picric acid. The non-poisonous coloviring-matters are some of the aniline colours, so long as they are pure, and contain no arsenic — saffron, turmeric, annatto, chlorophyll, and generally (with some exceptions) all organic colours obtained from the vegetable and animal kingdoms. The first thing for the analyst to ascertain is whether the colouring material is insoluble or soluble in water, for, as a rule, with the exception of gamboge, the harmless colours are soluble, while the mineral are insoluble in water. The organic colours are also bleached by a solution of hypochlorite of soda. The aniline colours are soluble in alcohol. The search for poisonous matters more properly belongs to, and is treated of, in the author's work on "Poisons." With the ■exception of salts of lead and copper in small quantities, they are rarely met with in food, and even in the matter of confectionery, of late years, there has been a great improvement. As a rule, sweetmeats in England ai-e not coloured with injurious matters.* The analyst having settled that the colouring-matter is one of organic origin, by its being bleached by sodic-hypochlorite, and by its solubility in water or alcohol, will next jn-oceed to study its spectroscopic characters, either by using a pocket spectroscope, or the micro-spectroscope already described. Mr. Sorby makes use for his instrument of little cells, cut from barometer tubing. They are half-an-inch long, and with an external diameter of somewhat under half-an-inch; they are ground flat at each end, and cemented with Canada balsam near one edge of a glass plate, so that they may be examined sideways or endways. In examining an unknown colouring-mattei", he adopts the following divisions: — 1. Soluble in water, and not precipitated by alcohol. 2. Soluble in water, but precipitated by alcohol. 3. Insoluble in water, but sohible in alcohol. * This is the more necessary to state clearly, since, on the Continent, very erroneous ideas prevail. Thus, in the Dictionnaire des Alterations et Falsifica- tions, parM. A. Chevallier et M. Fr. Baudrimont, Paris, 1878, the adultera- tions of half a century ago are enumerated; and the reader is informed that the English confectioners not onlj' falsify their sweetmeats with plaster, lime, starch, baryta, but frequently employ bronze powder, the leaf foil of copper, tin, and carbonate and arsenite of copper, verdigris, chromate of lead, red lead, and vermilion ; and, further, that nearly all the ginger lozenges con- tain lead. Similarly, in Dr. Hermann Kleucke's Lexicon der Verfdlschungen, in the article, " Conditorwaaren,'" it is stated that almost aU the English confectionery contains lead salts, often to the extent of one and a-half per cent. ! ! All this is nonsense. Such adulterations have been found, it is true; but instead of being common, they are rare and exceptional. 92 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ oD. 4. Insoluble in water and alcoliol. He next subdivides his main divisions accoi'ding to the action of bisulphite of soda. The organic colouring-matters most likely to be found may- be treated of in the order of the spectrum, beginning with the red. § 59. Eeds. — The common reds are — cochineal, aniline reds, alkanet, and the madder-colours alizarine and purpurine. Cochineal. — Cochineal is a red complex colouring - matter, secreted by certain species of a peculiar family of insects feeding on the Cactus coccinellifera, C. opuiitia, C. tuna, C. pereskia. The chief colouring-matter of cochineal is " carminic acid," the formula of which appears to be C^^HjgOjo. By the action of dilute acids carminic acid splits up into sugar, and a beautiful colour known as carmine-red, thus — Carminic acid. Water. Carmine-red. Glucose. CuHisOio + 2H2O - CiiHioOy + CeHioOs. Cochineal imparts its colouring-matters both to alcohol and water, and is precipitated by acetate of lead, carminate of lead being one of the constituent parts of the precipitate. The solu- tions of cochineal are purplish-red to crimson, turning a more or less rich violet-purple with alkalies, and becoming of a yellow colour on the addition of acids. The colour is well-known to chemists, as it is much used as an indicator for acids, being especially useful in titrating an alkaline liquid containing car- bonates, since carminic acid is not affected by carbon dioxide like so many other colouring-matters. Cochineal in neutral solutions gives absorption-bands, but not very definite when examined by the spectroscope ; if, however, it be made ammoniacal, then there are bands which difier in position only slightly from the absorption-bands of blood. No. 18 (fig. 16) is a graphical illustration of the spectrum of cochineal in water; No. 19, in alcohol; and No. 20, on the addition of nitric acid {a.) or NH3 (b.). If alum is added to cochineal it loses its power of turning yellow with acids, and the purpurine band becomes so bi'oad that the two bands almost run into each other. On addition of acetic acid they are separated, and appear as tolerably sharply-defined bands between J) and E, and there is another at D. On dissolving cochineal with alum solution, a lake is obtained j on dissolving this in tartaric acid, or dilute nitric acid, the solu- tion gives a band at b and E, and another close on D. The nitric acid solution gives a spectrum very similar to blood. 59.] COLOURING-MATTERS. 93 An aqueous solution of cocliineal may be distinguished from the red solutions of brazil-wood, sapan-wood, peach-wood, and a few others, by the fact that the calcium salt of their colour- ing-matters is violet, and readily soluble in water, while the calcium salt of cochineal-red is dark-purple or almost black, and insoluble in watei-. The following are the absorption factors for cartninic acid, as obtained by using a solution containing 0-0001 grm. per cc. with a drop of NHg : — * At Temp 20-0°. Absorption ractoi-. X 627-5 to X 609-5 0-00113 X5S3-6 „ X 576-6 0-00015 X 552-6 „ X 540-6 0-00010 X 5-22-7 „ X 51.3-2 0-OOOOS X 501-6 „ X 493-6 0-0001 X 482-8 „ X 475-0 0-00014 Aniline Reds. — The aniline reds are numerous ; the chief are fuchsine, safranine, and coralline. These three may be roughly distinguished from each other by adding a dilute mineral acid : fuchsine becomes yellow, safranine violet-blue, and coralline gives a yellow precipitate. Fuchsine, or Rosaniline, also called magenta, aniline red, and other names. It is a mixture of hydrochloride or acetate of para-rosaniline (triamido-triphenyl-carbinol) and rosaniline (tri- amido-diphenyl-tolyl-carbinol). It is distinguished from coral- line, which gives a very similar spectrum (see No. 28) by the yellow colour with acids already mentioned. The absorption spectrum of fuchsine has been studied by many observers, among others by Yierordtf and by Hartley. J In weak solutions (0-024 mgrm. per cc.) it shows an absorp- tion band in the visible spectrum when viewed through a stratum 4 mm. thick, extending from X546-7 to a.535-0 ; in stronger solution, 0-12 mgrm. per cc, and viewed in a layer * Ueber den Einfluss der Temp, gefiirbter Losungen u. die Absorption Spectren derselbeii zur ([uantitativen Spektralanalyse, von G. Kriiss u. H. Kriiss. Ztit. f. anorcjan. Chemie, i. + Op. cit. X The molecular structures of carbon compounds and their absorption spectra. Journ. Chem. Soc, li., 18S7. 94 FOODS : THEIR COMPOSITION AND ANALYSIS. L§ 59. 4 mm. in thickness, the band in the visible spectrum occupies the region from X462 to X580 ; there are also two other bands in the ultra violet — viz., one from X300 to X283, and another from X247 to X231, both, of course, invisible save with a Soret's ocular or other similar arrangement. A solution of rosaniline hydrochloride in alcohol (0-155 mgrm. per cc.) in a layer of 5 mm., gives an absorption in the visible spectrum from X591 to XioO, and in the ultra violet from X310 to X'274:. Fig. 16. The absorption factor for the spectrum, from a to H, has been worked out by Vierordt, and, reducing his notation to wave lengths, the following are the results from B to F : — EXPLANATION OF FIG. 16. 95 No. 1. A diagrammatic re])resentation of tlie bands of No. 2. 2. Absorption spectra of clear blue cobalt glass. 3. Absorption spectra of dark blue cobalt glass. 4. Diagrammatic representation of No. 3. 5. Alcoholic solution of alizarine. G. Alcoholic solution of alizarine made alkaline with ammonia. 7. Aqueous ammoniacal solution of alizarine. 8. Alcoholic solution of alizarine made alkaline by potash. 9. Sulphoxanthraquinone in alcoholic solution alkalised by potash. 10. Aqueous solution of sulphoxanthraquinone made alkaline by potash. 11. Alcoholic solution of purpurine. 12. The same alkalised by ammonia. 13. The same alkalised by potash. 14. A neutral solution of alizarinamid. 15. An ammoniacal solution of alizarinamid. 16. A neutral solution of purpurinamid. 17. The same alkalised by baryta. 18. An a([ueous solution of cochineal. 19. A dilute alcoholic solution of cochineal. 20. Cochineal in concentrated watery solution with (a.) nitric acid, (b.) ammonia. 21. Logwood, (a. ) concentrated watery solution, (6.) dilute. 22. The same with the addition of nitric acid. . . . The same alkalised by ammonia. 23. A decoction of brazil-wood. 24. The same alkalised by ammonia. 25. Litmus, {a.) concentrated, {b.) dilute. 26. The same made acid, (a.) concentrated, (b.) dilute. 27. Dilute solution of fuchsine. 28. Alcoholic solution of coralline. 29. Alcoholic solution of eosin, (a.) concentrated, (6.) dilute. 30. Safranine, [a.) concentrated, (b.) dilute. 31. Naphthaline red, (a) concentrated, (b.) dilute. 32. Curcuma, (a.) concentrated, (b) dilute, (c.) strongly diluted. 33. Fluoresceine, (a) somewhat concentrated, [b.) dilute. 34. Fustic extract. 35. Fresh chlorophyll in alcoholic solution. 36. Old chlorophyll solution. 37. Wine colouring-matter, (I.) pure, (II.) diluted. 38. Wine colouring-matter + NH3. 39. (I.) Alallow colouring-matter concentrated, (II.) Elderberry con- centrated. 40. Acid cherry : — (&.) Acid cherry, with addition of tannin. 41. Mallow colouring-matter, with the addition of alum. 42. Indigo solution. 96 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 59. Brighter portion of the Spectrum extending from b nearly to d. Wave Lengths. Absorption Tactor. 679-9 to 663-2, -01447 663-2 ,,646-2, -00449 646-2 „ 629-4, . . . ; . . '00207 629-4 ,, 612-6, -00117 612-6 ,, 595-9, -000517 Darker portion of the Spectrum from about D to F. 595-9 to 5S2-3, -000127 582-3 ,,569-9, '0000592 569-9 „ 557-0, -00002238 557-0 ,, 546-8, -00001248 546-8 „ 535-0, -000007819 535-0 ,,523-6, -000009669 523-6 „ 516-3, -00001222 516-3 ,, 508-5, -00001376 508-5 „ 50ri, -00001394 501-1 ,, 494-2 -0000153 494-2 „ 486-0, '0000176 Safranine dissolves in alcohol with a fine rose-red colour, with a weak red fluorescence. Its spectrum is shown in No. 30 {a, concentrated solution, h, dilute) ; it is like that of eosin (No. 29), but the spectrum of eosin is changed by nitric acid, that of safranine is unchanged. From a solution containing safranine, safranine ma}^ be extracted by shaking it up with its own volume of amyl-alcohol in the tube already described, page 69. The spectrum of naphthaline-red is figured No. 31, after "Vogel, but it is not quite accurate, for it gives a delicate shaded band from X536 to X558, and a band extending from E to X472. Coralline and Aurine (rosolic acid) give a very similar spectrum. The spectrum of aurine, according to Noel Hartley, in 0-00058 grm. per cc, viewed in a layer of 3 mm., consists of a band, X516 to A361, in the visible, and also a band in the ultra violet spectrum, X286 to X256 ; in weaker solution (0-000116 grm.) there is a single band, X484 to X467. Fonceau 2R (C^gH^^NgO-SoNa^), the sodium salt of xylene- azo-/3-naphthol-disulphonic acid ; Ponceau 3 R, the similar salt of cumene ; Congo-red^ CgoHooN^jOi^S.jNa.,, the sodium salt of diphenyl- diazo-binaphthionic acid, all give a similar spectrum, a broad band in the green, from about X448 to X488. Erythrosine (phloxine), CogHgOjI^Na^, the sodium (or potas- sium) salt of tetra-iodo-fluoresceine, gives two bands in dilute solution, the one extending from X488 to X530, then a narrower band between D and E, a550 to X558. § 59.] COLOURING-MATTERS, 97 Fast Red, CooHjglSToO^SNa, the sodium salt of ^-sulplio- naphthalene-azo-./S-uaphtho], gives in solutions of 0-04 mgrm. per 100 cc. viewed through 1 mm., a band between F and G, X4:41 to X439. Biehrich Scarlet, Cc,.-^-^^^ fi-'^.^a..^, the sodium salt of sulpho- benzene-azo-sul]ihobenzene-azo-/3-naphthol, gives in a solution, 5 mm. thick, strength 0-56 mgrm. per 100 ce., a baud extending from X518 to X438 {Hartley). Croceine Scarlet, Oo4Hjg]Sr^07SoN'ao, the sodium salt of sulpho- toluene-azo-toluene-azo-/3-naphthol-/3-sulphonic acid, gives in a solution, 2 mm. thick, strength 0-45 mgrm. per 100 cc, a band from X520 to X479. Alkanet, the root of Anchusa tinctoria, contains a red colouring- matter, insoluble in water, but soluble in alkalies, alcohol, ether, and fatty oils. The colouring-matter appears to be an acid, anchusic acid, Cg^H^pOg. In dilute solutions the spectrum shows three absorption-bands; on the addition of a trace of a magnesium salt, a fourth absorption-band appears, hence alkanet-red is a test for magnesium salts, and conversely a magnesium salt is a test for alkanet-red.* It may not unfrequently be found as the colouring-matters of tooth tinctures, hair oil, &c. Madder. — The root of the madder, Ruhia tinctorium, contains two colouring -matters — alizarine and purpurine — with others less studied. Alizarine, Cj^HgO^, crystallises from an alcoholic solution in yellowish-red crystals, and may be sublimed as brilliant red needles at temperatures a little above 100°. The needles are sparingly soluble in water, but dissolve freely in alcohol and ether. Alizarine is now made artificially on a large scale. The alcoholic solution of artificial alizarine shows no bands, but there is extinction of the violet up to the green (see No. 5). On the addition of ammonia the solution changes to a beautiful red, with weak bands in the green (No. 6) ; the aqueous ammoniacal solution gives two bands (No. 7) ; the alcoholic solution made alkaline with potash gives evidence of a third feeble band (No. 8). Natural alizarine is not now much used, but it may be at once distiuguished from the artificial by its giving absorption-bands in an alcoholic solution. Alizarine may be also distinguished by chemical tests : copper acetate added to a solution in alcohol gives a purple precipitate, aluminium acetate gives a red precipitate in an alkaline solu- * Magnesium salts also alter more or less characteristically the spectrum of juice of elder berries, the colouring-matters of the beet, dahlia, dragon's mouth, horse chestnut, hyacinth, violet, rhododendron, purple aster, and Primula farinosa. Bericht der deutuch. Gesdlschajt, xiii., 7Co-768. 8 98 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 59. tion, azid ferrous acetate and other iron salts give a dark blue violet precipitate. Blue precipitates are also formed on the addition of either barium chloride or calcium chloride to alkaline solutions. These precipitates have respectively the composition Ci^HgO^Ba and Ci.Hp^Ga. Purpurine, Cj^HgOr,, crystallising from an alcoholic solution in yellow needles, and subliming like alizarine in red needles, dissolves in alkalies with a dark red colour ; it gives not blue but purple precipitates with the chlorides of calcium or barium. The alcoholic solution gives (see No. 11) weak absorption-bands at F and at b E. The spectrum on the addition of ammonia or potash becomes very characteristic (see Nos. 12 and 13). Saffloicer. — The safflower, Carthamus tinctoruis, contains in its petals several colouring-matters, chief among which is carthamine or carthamic acid, C^^H-^QO►■. Carthamine turns red in alkaline solutions, and may be precipitated red by an acid. It is met with usually as a delicate pink or red dye, and forms the usual colouring-matter of rouge. Logivood contains a colouring -matter named hnematoxylin, Cj(;H-^^Oq + SHoO, crystallising from water in yellow prisms ; this changes by the action of the air and ammonia into a red or purple substance named hcematein, (OjgH^^gOg),^". It may also be produced by adding nitric acid to an ethereal solution of hsematoxylin. Hcematein always exists in the free state in the wood, and an alcoholic tincture gives the reaction of hgematein. The spectrum of the logwood colouring-matters is delineated in Nos. 21 and 22. Alkalies turn the tincture first red, and then violet. The best test, however, is the addition of an aluminous- salt to an ammoniacal tincture of logwood, the result being the formation of an abundant bluish-violet precipitate. To test for alum, chips of logwood are boiled with water and the solution diluted until the spec- trum shows the edge of the band close to the D line, as represented in fig. 16, curve 22; Fig. 17. on now adding a drop of a solution of alum, 1 per cent., the absorption in the blue disappears, and a single band makes its appearance, having its centre about the line D, and extending from E to between C and D, or to about X628 — the curve then having the position shown in fig. 17:1 mgrm. of either alum or Al^Cl,, may be in this way recognised. Iron interferes with the test, and should it be present, Vogel recom- mends its conversion into the ferrous condition, to add ammonium § 59.] COLOURIXG-MATTERS. 99 sulpliocyanide, and to extract the sulphocyanide with ether, iron sulphocyanide being soluble in ether ; in this way he has been able to detect alum in a mixture of 1 of alum and 40 of an iron salt {p2). cit.). Coloured liquids, such as wine, are decolorised by a little Fau de Javelle and hydrochloric acid, boiled with alcohol, neutralised, and filtered. Brazil-wood sapan-wood, lima-wood, peach wood, and some others, yield a glucoside, which splits up into sugar and brasiline, CooHoqO-. Alkalies turn brasiline a crimson colour, and the crimson solution gives blackish-violet precipitates with alu- minium and stannic salts. It is oxidised slowly in the air, or rapidly by the action of nitric acid into a red crystalline body, hrasilein, (C22Hjr,07)3N. The spectrum of brazil-wood extract in solution is delineated in Nos. 23 and 24, There is also a red coloux'ing-matter yielded by the santal- wood, named " santalin," and occurring in microscopic red crystals insoluble in water, but dissolving in alcohol, with a red colour, turning to violet on the addition of alkalies. The lied Juices of Fruits and Berries. — Red currants when, fresh show a band between D and F, the centre of the band being nearly at E. In dilute solutions there are indications that this is really a double band ; when acidified, the colour can be shaken out with amyl alcohol. The red colouring-matter of the strawberry is also similar, if not identical. It gives a double band in the green, and the colour is dissolved from the acid solution by amyl alcohol. The raspberry has also a similar, if not identical, colouring- matter. The spectrum of these vegetable juices appears to be analogous to the red in the cromatophore of the lobster {Thudichum). The red berries of the common asparagus contain a yellow colour soluble in water, and a red colouring-matter soluble in alcohol. The latter colouring-matter gives two well-defined bands between F and G. The red berries of Crataegus also give a two-banded spectrum, the one between E and F, the other midway between F and G. All these colouring-matters, however, require farther investi- gation, especially with regai'd to the ultra-violet and infra-red portions of the spectrum, both of which are invisible. The infra-red spectrum will specially repay study. A method of examination of this portion has lately been discovered of great accuracy. A fine wire, through whicJi a feeble current of elec- trical current passes, is slowly, by means of a delicate micro- meter arrangement, moved along the invisible spectrum, and where there are absorption bands or lines the electrical current 100 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 60. is affected as measured and shown by a galvanometer. The ultra- violet part can, to some extent, be made visible by photography, but tlie easier process is, as before mentioned, by the use of a Soret's ocular. The red colouring-matters of the grape and various reds used to colour wines are dealt with specially in the article on Wine. As with other colours, so with red tints generally ; if they are once identified, imitation mixtures can be made and esti- mated colorimetrically ; or, if the absorption factor is known, they may be estimated by a spectroscope with a double slit and a suitable shutter connected with the ocular. § 60. Orange and Yellow. — The most common oranges and yellows are the annatto colours : curcuma, picric acid, fustic, chrysophanic acid, gamboge, and aniline oranges and yellows. The Annatto Colours. — The colouring-matters of the annatto are two substances, one a yellow, orelliii ; another a cinnabar red substance, bixin. The latter is described in the article on " Annatto." The annatto colouring-matters are not soluble in water, but are easily dissolved by alcohol. The alcoholic solution is orange red, and non-fluorescent. On the addition of nitric acid it becomes turbid. On dilution with water, there is a strong fluorescence, and it becomes yellow-green. It then absorbs, like ferric chloride, the whole left side of the spectrum E, and half to D. Turmeric is the root of Curcuma longa. The colouring-matter is curcumin, CgH^gOg, insoluble in cold water, and sparingly soluble in boiling water. It is very soluble in alcohol, and forms brilliant yellow crystals. Turmeric moistened with boric acid and dried assumes an orange colour, changed by alkalies into a blue ; this is due to the formation of a compound soluble in alcohol, forming a red solution, and crystallising in lustrous green crystals, to which the name of rosocyanin has been given. No. 32 shows the spectroscojoic appearances of curcumin. Picric Acid (CgH3(NOo)oO), also called carbazotic acid and irinitrophenol, is formed commercially by acting on phenol, by dissolving it in sulphuric acid, and then treating the solution with nitric acid. It ciystallises from hot water in yellow plates, having a very bitter taste. The salts are explosive. It is taken up from acid watery solvitions by petroleum-ether, ether, or benzine, and hence can be readily obtained pure enough for examination. Picric acid is not precipitated by acetate of lead. The chief chemical test is the production of isopurpurate of potash, whicli is the result of adding cyanide of potassium, and gently •warming. § 60.] COLOURING-MATTERS. 101 The reaction is represented as follows : — C6H3(N02)30 + 3CNK + 2H2O = C8H4N5O6K + CO3K2 + N H3. Isopurpurate of potash is of a blood-red colour. Fustic is the general name for yellow coloiirs found in the wood of the Morus tinctoria. The wood contains two distinct colouring-matters ; the one, moritannic acid, Cj3HjgOg + HgO, soluble in hot water, and form- ing yellow crystals ; the other, morine, C^oHgOg, is but sparingly soluble in water ; crystallised from alcohol the substance forms yellow needles. Both give yellow precipitates with- acetate of lead. The spectroscopic appearances of fustic are shown in No. 34 (a.) in concentrated, ib.) in dilute solution. Chrysophanic Acid, C^jH^qO^, appears to be dioxymethylan- thraquinone. It is contained in the rhubarb and wall lichen {Parmelia jxcrietina). In commerce it is in the form of six-sided tabular crystals, of a pale to an orange-yellow colour. It is very readily extracted from rhubarb which has been pre- viously macerated in water, pressed, and dried, by digesting the rhubarb in benzine. The crystals are soluble in ether, oil of turpentine, coal-naphtha, benzine, and other hydrocarbons. The spectrum of chrysophanic acid is very similar to that of natural alizarine. Solutions of chrysophanic acid give, with alkalies, a rich purple colour. An ammoniacal solution of chry- sophanic acid yields, with alum, a beautiful rose-coloured preci- pitate. An alcoholic solution of subacetate of lead gives, in the alcoholic solution, a red-white precipitate. Chrysophanic acid, on being oxidised with nitric acid, yields methylanthraquinone. Gamboge.- — This is a colouring-matter derived from the Garcinia morella, var. pedicellata. It is a gum resin, to which the formula of CyoHggOg has been ascribed. It is very insoluble in water, but soluble in alcohol. In order to detect it in sweetmeats, to which it has occasionally been added, the yellow colouring-matter is dissolved in alcohol, and precipitated by water, a reaction at once showing that it is of a resinous nature. The precipitate, if gamboge, will give a red colour with an alkali. It is without doubt a poison- ous colour, but the question of quantity must be considered. Sixty grains have caused death, and five grains is a medicinal dose; and since some persons, especially children, are peculiaidy susceptible to the action of medicines, even quantities so small as a third of a grain in a couple of ounces of sweets, should be considered as having possibly an injurious action. 102 FOODS : THEIE COMPOSITION AND ANALYSIS. [§ 60. Aniline Oranges and Yellows. — Eosine, C.2QHgOr,Br^Na.2, the alkali salt of tetra-bromo-fluoresceine, is a reddish-brown, powder soluble in water, the solution being of a reddish-orange with green fluorescence; it gives a yellow-red colour, or, if in strong solution, a red precipitate with hydrochloric acid. All the derivatives of fluoresceine have a similar spectrum, of which eosine may be taken as the type. Moderately diluted, there is an intense dark band in the green; this, farther diluted, divides into two well marked bands, the respective centres of which (in an alcoholic solution) are X525 and X491 ; the same bands, slightly altered in position, are seen in ethyl-eosine, rose bengal, and phloxin. Eosine as a dye may often be recognised by the spectrum of the fabric strongly lit by the reflected rays of the sun, or by the lime or electric light. The fibre may also be digested in slightly acidified alcohol ; the alcoholic extract is then taken up by amyl alcohol, a drop of ammonia added, and the specti-oscopic appearances noted. The absorption factors of eosine from a solution of O'Ol mgrm. per cc. have been ascer- tained by Kriiss for the following wave lengtlis : — Temp. 20°. Wave Length. Aljsorption Factor. 678-9 to 656-2 0-00025 627-5,, 609-5 . ■000-20 583-6 „ 576-6 . 0-00012 552-6 „ 540-6 . 0-0000.55 522-7 ,, 513-2 . 0-0000096 501-6,, 493-6 . 0-0000137 482-S „ 475-0 . 0-0000221 46ii-0 „ 461-5 . 0-0000344 Croceine Orange (Ponceau 4), C^gHj^NoO^SNa, is the sodium salt of benzene-azo-/3-naphthol-/3-sulphonic acid. This dye is soluble in water with an orange colour. Hydrochloric acid gives a brownish-yellow precipitate. It gives a spectrum in which a single broad band occupies half of the green and some of the blue ; the band in dilute solution extends from about X526 to a438. Orange T (Mandarin G R), C^yH^glSroO^SlS'a, is the sodium salt of sulpho-o-toluene-azo-/3-naphthol, it is soluble in water, and precipitated by hydrochloric acid in brown fiocks. It gives a similar spectrum to the above. Orange I (Tropreoline 000), CjgHj^NoO^SNa, is the sodium salt of ^>2ulpho-benzene-azo-a-naphthol, while Orange II is the similar compound of /3-naphthol ; the former gives a brownish- yellow colour to hydrochloric acid, and precipitates in brown flocks. It gives a spectrum similar to croceine orange: 0-2 mgrm. § 60.] -COLOURING-MATTERS. 103 in 1 cc, viewed in a stratum of 4 mm., gives a band extending from xr)30 to x:399. Resorcin Yellow (Tropseoline O), OjoHpN^OjiSXa, is the sodium salt of 7>sulplio-benzene-azo-resorcinol ; it is soluble in water, and not changed by hydrochloric acid ; dissolved in the pi'opor- tion of 0-2 mgrm. in a cc. of water, and viewed in a stratum of 4 mm., it gives a single band from X475 t?o X361 {Hartle;/). Biphenylamine Orange (Orange IV — Tropteoline 00 — Fast yellow), CjgH-^^NgOgSNa, is the sodium salt of p-sulpho-benzene- azo-diphenylamine. The salt dissolves in water readily ; it gives a violet precipitate with hydrochloric acid. It gives no bands, but cuts off the violet and blue rays. Chrysoidine, CjoH^gN^CL, is the hydrochloride of diamido-azo- benzene. In commerce it is more often mixed with other colours ; thus, mixed with magenta, it is known as " cardinal ; " with safranine, it is known as " scarlet for cotton ; " it is soluble in water, and strikes with hydrochloric acid an orange-brown colour. It gives a band in the visible spectrum, when in the concentration of 0'8 mgrm. per cc, and viewed through 2 mm., extending from X494 to X439 ; in a stratum of 3 mm. there is a band in the ultra violet from X247 to X232 (^Hartley). Methyl Orange (Orange III— Helianthine), Ci^H^^NgSOglSra, is the sodium salt of ;>sulpho-benzene-azo-dimethylaniline ; it is a common laboratory reagent as an indicator for alkalies and acids. It is readily soluble in water to an orange solution, and strikes a magenta red with hydrochloric acid and other acids. Methyl orange in a concentration equal to 0-3 mgrm. per cc, and viewed through a stratum of 3 mm., gives a band extending from X509 to X361 ; it, therefore, shuts off all the violet and blue rays, as well as some of the green. It also, under the same conditions, gives a spectrum in the ulti-a violet, extending from a270 to X256. In a more dilute solution (O'OG mgrm. per cc.) it gives a band from X458 to X407 {Hartley). Metanil Yellow, C^gHj^NgOoSNa, is the sodium salt of m-sulpho- benzene-azo-diphenylamine ; it is soluble in water, and gives a crimson colour with hydrochloric acid. It gives no band in the visible spectrum, but shuts off all the violet and blue rays. FJiosphine (Chrysaniliiie), CjgHjgN'^Og, is the nitrate of diamido-phenyl-acridine and its homologues. This colour is soluble in water, and does not change colour with hydrochloric acid. It gives a shadowy band from a486 to a508 in the visible spectrum. 104 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 61- § 61. The Greens — are cliloropliyll and the aniline greens. Chlorojihyll, the green colouring-matter of plants. It would appear that it is now possible to separate chlorophyll in a crystal- line state. The merit of this discovery belongs to A. Gautier, who crystallised chlorophyll in 1877,* and two years afterwards IToppe-Seyler,t without knowing of Gautier's paper, described the same substance under the name of chlorophyllone. Gautier's method is as follows : — The leaves are pounded in a mortar, and sodic carbonate added in quantity nearly sufficient to neutralise the acid juices, and the product strongly pressed. The marc is then exhausted at 55°, and again strongly pressed ; to the ex- hausted substance alcohol is added, and the whole digested at 83° ; chlorophyll, wax, pigments, fat, and other matters dissolve. The liquid is filtered and digested with coarse animal charcoal, which has been previously washed. At the end of fou.r or five days it has lost its green colour, and becomes yellowish- or brownish-green. The charcoal retains the chlorophyll, and is now washed with alcohol at 65° ; the latter removes a yellow crystallisable substance generally accompanying chlorophyll, and intimately allied to it in composition. From the animal black thus freed from the yellow substance, chlorophyll may be extracted by anhydrous ether, or very light peti'oleum ether. A slow evaporation in the dark wall yield it in: crystals. Thus obtained, chlorophyll forms flat, often radiating crystals, which may be more than a centimetre in length, soft in consistence, and of an intense green colour when recent, but slowly changing to yellowish-green, or greenish-brown. If the crystalli- sation is too rapid, these long crystals are not obtained, but green masses composed of microscopical crystals, belonging to the ]'hombohedral system. Difl\ised light changes chlorophyll to yellowish-green, and ulti- mately decolourises it. The crystals dissolve in ether, chloro- form, petroleum, carbon disulphide, and benzine. Digested with hydrochloric acid, chlorophyll splits up into two new substances, one giving a beautiful blue solution, the other remaining insoluble, but dissolving with a brown colour in hot ether and alcohol, from which it appears inclined to crystallise. Fremy, who was the first to notice this splitting-up of chlorophyll, called the first substance phyllocyanine, and the last phylloxanthine. The ultimate analyses of chlorophyll by Hoppe-Seyler and Gautier agree fairly well, especially as Hoppe-Seyler' s chlorophyli being derived from monocotyledonous and Gautier's from dico- * Bulletin de la socldte cliem. T. xxviii., 1877, p. 147. Comptes rendus, Ixxxix., p. S61. + Bericht der deutschen chem. Gesellschaft, 1879. § 61.] COLOURING-MATTERS. 105 tyledonous plants, it is possible there may he some slight difference in their composition. Chlorophyll has not yet been obtained free from ash. Hoppe-Seyler. Gautier Carbon, . 73-4 73-77 Hydrogen, 9-70 9-SO Nitrogen, . 5-62 415 Phosphorus, 1-37 Phosphates, Ash, 1-75 Magnesia, . '. -34 Oxygen, . 9-57 10-33 Mr. Sorby has studied the colouring-matters of leaves, and has divided them into five groups : — 1, Chlorophyll ; 2, xanthophyll ; 3, erythrophyll ; 4, chrysophyll ; and 5, phaiophyll. The Chlorophijll group are all green in colour; the members are insoluble in water, but soluble in alcohol or carbon disulj^hide. The absorption-band is in the red, biit the green is more or less completely transmitted, so that the prevailing tint is a more or less modified green. There appear to be several members of the group, one kind (which is probably the ci-ystalline chlorophyll just described) occurs nearly pure in small aquatic plants, allied to the Oscillatoria ; the green leaves of plants contain this, along with one which gives special absorption-bands. A third kind is the result of the action of acids on this, found more especially in autumnal leaves which have become brown. A fourth is in faded Confervje, and turns blue when acted upon by hydrochloi'ic acid (Fremy's phyllocyanine ?) Xantliopliyll. — This is a group of orange colours. They are insoluble in watei", but soluble in alcohol or carbon disulphide. The general tint of the spectrum is orange, there is absorption at the blue end, often more or less mai'ked nai'row bands. These bands are best seen in carbon disulphide solution. Examples of two species of xanthophyll are the inner and outer layers of the common carrot. Erythrophyll. — A group of reds. They are soluble in water and weak alcohol, but not in carbon disulphide. There are many varieties. There is a strong absorption in the green parts of the spectrum. (Jhrysopihyll is a golden-yellow colour, soluble in water and weak alcohol, like the last, and insoluble in carbon disulphide. The Fhaiophyll group consists of various browns, due to the oxidation of chlorophyll. They give no definite absorption- bands. The spectrum of chlorophyll is given in ISTos. 35 and 36. (fig. 16). No. 35 is an alcoholic solution of fresh chlorophyll. There are two well-defined absorption-bands in the red, and lOG FOODS : THEIR COMPOSITIOX AND ANALYSIS. [§ 62, two others, somewhat weaker, between E and D. The blue end •of the spectrum, up to F, is dark. Old solutions (No. 36) alter the bands somewdiat, and a fifth band appears between F and C Green, if not due to chlorophyll, nor to colouring-matters of ■copper or metallic origin, may be one of the aniline greens, such as aldehyde or iodine green. Aldehyde Green, O20H27N3S2O, is made by the action of aldehyde on rosaniline and treatment of the product with sodium hyposulphite. It is only manufactured in small quan- tities, its place having been taken by other greens. It is insoluble in water, but soluble in alcohol. The spectrum of aldehyde green, methylrosaniline picrate, iodine green, and malachite green are all similar, but by the aid of tests and comparative solutions, may be distinguished from each other, ■p- ,0 The following diagram (fig. 18), after Vogel, suffi<;iently indicates the differences between the spectra. Mr. Hartley has investigated the spectrum of iodine green in a concentration of 1-25 nigrm. per cc, viewed in a stratum of 5 mm. ; it gives the following bands :— In the visible part two bands, one from xeQl to X545, the other from XiM to X410 ; in the ultra violet a single band from X324 to a286 ; but in a layer of 4 mm. two bands are seen in the ultra violet, one from X6n to a286, the other from X256 to X235. Probably most, if not all, of the aniline greens give bands in the ultra violet. § 62. Indigos and Violets.- — The chief blues to be found, not of mineral origin, are indigo, litmus, and the aniline colours. Indigo. — Indigo is the produce of a great number of plants, most of which belong to different species oi Indogifera. It would seem that the indigo exists in these plants in the form of a syrupy glucoside, whicli has been named indican, and splits up into indiglucin and indigo-blue, according to the equation, Indican. AVater. Indigo-blue. Indiglucin. cCh2^4 + 4H2O = C16H10N2O2 -f 6C6h7A;- In the treatment of the crude substance by acids, other red and yellow colouring-matters are thrown down with the indigo. Pure indigotin has to be freed from these impurities, and may be crystallised in minute crystals. It is also capable of sublimation. Crystalline indigotin is of a deep purple colour, and is insoluble in water, dilute acids, and alkalies. It is converted into orange §6: COLOURIXG-MATTERS. 107 agents. Indigotin is dissolved in concentrated sulphuric acid, and two compounds are formed — the one, indigo monosuljjhonic acid, or indigo-purpuric acid, Cj(5HnNoO„(S03H), which is first formed ; and theother indigo disulplionicacid, Cj(5HglSroOo(S03H)2, which is formed later. The first is easily separated from the second, for it is precipitated by the addition of water. A sodium salt of indigo-sulphuric acid is used much in laboratories under the name of indigo-carmine as a test for nitrates, as well as for their estimation, oxidising agents converting it into isatin- sulphonic acid, C;[gHgIS]'o0j(S0oH)2, which has an orange colour; hence, when a liquid containing a nitrate is acidified strongly with sulphuric acid, and a solution of indigo-carmine run into the hot solution, the colour changes at first to yellow until an excess of indigo has been added. The spectrum of indigo, in concentrated solution, is shown in No. 42. There are no bands, but absorption of the red, orange, and yellow end of the spec- trum. Indigo can be determined quantitatively by the spectro- scope ; the absorption factor for a412'7 to X595-9 is 0-0000142. Litmus is a blue colour obtained from lichens, and very familiar to chemists as an indicator of acids. As a colouring agent of either foods or stuffs, it is of little importance, not being a " fast colour." Its spectrum is shown at Nos. 25 and 26. Aniline Bines. — The aniline blues and violets are numerous, and give spectra in which may be seen one or more bands in the red. Methylene Bine, CjoHisNoSCl or (Ci6H^sN3SCl)o + ZnCU + H.O, is the hydrochloride, or may be found as the zinc double chloride of tetra-methyl thionine. It is easily soluble in water, but less soluble in alcohol. Reducing agents reduce it to leuco- methylene blue, the air oxidising it back again to methylene blue. Concentrated solutions shut oft" all red and yellow up to >.547; more dilute ones show a band from D to C (X589 to a622). The absorption factor has been worked out for certain wave lengths.* Temp. 20°. Wave Lengths. Absorption Factor. X678-9 to \656-2 0-0001996 0-000265 0-000725 0-00125 0-00308 00326 0-005 0-0176 Mauve (Perkins' Violet), the colouring matter of postage stamps, is soluble in water, and gives a band from X614 to X558. * G. and H. Kriiss, Zeit. f. anorgan. Chemie, I. X627-5 to \609-5 X583-6 to \576-6 X552-6 to \540-6 X522-7 to \513-2 X501-6 toX493-6 X482-8 toX475 X469-0 toX461-5 108 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 63. Aniline Blue (soluble in spirit) is the hydrochloride, sulphate or acetate of triphenyl-rosaniline and triphenyl-para-rosaniline. These blues are insoluble in water; properly diluted they give a band extending from d to D. Methyl Violet gives a band the centre of which occupies the line D. Alkali Blue 6 B gives no band, but extinguishes the red and yellow; on acidifying with HCl a band is produced between E and D, and there is absorption of the violet and blue. § 63. Brown C o 1.0 v rs. — Bismarck Brown (Vesuvine), Cj8H2Q]SrgCl2, is the hydrochloride of benzene-diazo-phenylene- diamine. It is soluble in water, and does not change by acidi- fying with hydrochloric acid. Hartley has investigated its spectrum. In a stratum 5 mm. thick, with a concentration 0-09 mgrm. per cc, it gives a band extending from ?.467 to X407. Acid Brotvn, O^gH^jNgOgSISra, is the sodium salt of benzene- azo-phenylene-diamine-azo-benzene-;>sulphonic acid ; it is soluble in water, and does not change by the addition of hydrochloric acid. A solution of 4 mgrms. per cc, viewed in a stratum 8 mm. thick, gives an absorption band from >.451 to a399. Caramel. — All shades of brown, from a rather delicate fawn colour up to black or nearly so, can be given by caramel. Carar mel is not a single simple substance, but a mixture of various colouring-matters derived from the dehydration of sugar. Cai'a- mel is soluble in water, and is precipitated by alcohol. Caramel, as it usually occurs, is almost all soluble in water, and gives precipitates with alcohol, ammoniacal lead acetate, and baryta water. It gives a spectrum without definite absorption- bands, but extinguishes the blue side of the spectrum, like ferric chloride. Caramel may be suspected in a brown liquid not coloured by a mineral substance, which is not decolorised when sufficient tannin is added, for the colouring-matter of berries is precipitated by tannin. Among the members of caramel, there are three principal substances — viz., caramelane, C^.iHoj.Og, cara- meline, and caramelin. Caramelane makes up the chief bulk of ordinary caramel. It is brown, hard, and brittle, bitter to the taste, without odour, deliquescent, and very soluble in water, sparingly soluble in weak alcohol, insoluble in ether. It does not precipitate metallic salts in aqueous solutions, but reduces an alkaline solution of oxide of copper, and also the salts of gold and silver. Carameline may be separated from caramelane by first extract- ing the latter by alcohol of 84 per cent., exhausting with cold water, and precipitating with absolute alcohol. It is a brittle solid, of great tinctorial power ; it is easily soluble in water, sparingly in alcohol, insoluble in ether. § 64,] COLOURING-MATTERS. 109 Caramelin is a mixture of three substances of different solubili- ties ; it is black, shining, and infusible. Its solutions precipitate metallic salts, and reduce gold and silver solutions. § 64.— SCHEME FOR THE DETECTION OF THE ANILINE AND OTHER COLOURING MATTERS* BY CHEMICAL REAGENTS. The aniline colours admit of a chemical classification as follows : — ■ I. Nitro- Colours. — The examples of which are picric acid, the duiitro- cresols, bromonitro-phenol-jo-sulphonic acid. IL Azoxy-Colours, such as the sodium salt of azoxystilbene disulphonic acid, known in commerce as " sun yellow." III. Hydrazine Colours, such as " tartrazines." IV. Azo-Colours. — These are very numerous. The bivalent azo-group \ — N = N — ) is found in two classes of organic bodies — in the diazo and in the azo compounds. In the diazo compounds the one valency is satisfied by a benzol residue, and the other by a halogen, hydroxyl, an inorganic acid, or an amido-group ; e.rj. — CeHs _ N = N — CI. (CgHs - N = N)o — SO4. Diazobenzol. Diazobenzol sulphate. CcHs - N = N - NHlCsHs). Diazoamidobenzol. CeHi — N = N — NO3 I C6H4 _ N = N — NO3. Tetrazodiphenyl-nitrate. These diazo bodies possess no colouring properties, but they are the more important sources of the azo bodies. The azo bodies contain two benzol residues, each of which is bound to a, Falency of the nitrogen atoms, thus — CeHs — N = N — CgHs. Azobenzol. CH3 CH3 C6H4 — N = N — CcH4. Azotoluol (ortho and para). C10H7 _ N = N - C10H7. Azonaphthalin. NOo CgH4 - N = N — CH2 — CO — CH3. Nitrophenylazoacetone. * The above classification is borrowed from The Systematic Survey of the Organic Colourimj- Matters, by Dr. G. Schultz and P. Julius, trans- lated, with additions, by Arthur G. Green, F.I.C., F.C.S., London, 1894. On the interesting chromophore theory of Witt, see Ber. der deutschen Chem. Ges., ix., p. 522, and Chemiker-Zeitwiri, iv., 422. On the poisonous properties of the tar colouring-matters, see Die Tlieerfarben, mit hesonderer Jiiicksicht auf Schadlichke.it u. Gesetzgebung, hygienischu.forensisch-chemlsch tmtersucht, von Dr. Th. Weyl. Berlin, 1889, 110 foods: their composition and analysis. [§64. The azo bodies possess colour, but are not dyes until an acid or basic group is introduced into the molecule ; azo-benzol, for example, is a yellow substance not capable of dyeing, but azo-benzol mono- or di-sulphonic acid possesses weak-dyeing properties. Similarly, if an hydroxyl or an amide group enters into tlie molecule, the dyeing properties are much enhanced j for example, amido-azo-benzol disulphonic acid — NH2(4) C6H4- - N = N - CgHs HS03 \, HS03, in the form of a sodium salt, is the chief constituent of " acid yellow ' tropseolLa Y is the sodium salt of oxy-azo-benzol sulphonic acid — CcH4 — N = N — CgH4(0H) About half the number of tar colours known belong to the azo-group. Examples of azo colours are : — Yellows. — Acid yellow, Sudans — croceine — orange — butter yellow. Reds. — Various Ponceaux, Bordeaux, Roccelline. Black. — Azonigrine — naphthylamine black D. PoRPLES AND Blues. — Hessian purple, diamine violet, sulphone-azurine. V. IN'itroso Colours. — All the colours in commerce belonging to this group are green ; examples are dinitroso-resorcin, naphthol green. VI. Oxyketone Colours. — Examples are alizarines — purpurine. VII. Diphenyl-methane Colours. — Examples, auramine, acridine red, pyronines. VIII. Diphenyl-methane Colours. — Examples, malachite green, Victoria^ blue, magenta, acid magenta, methyl violet, eosine, erythrosine, rose Bengal, phloxines, rhodamines. IX. Indophenols. X. Oxazines and Thiazines. — Examples are Meldola's blue (chloride of dimethyl -phenylammonium-i3-naphthoaxazine), methylene blue B and B G H (hydrochloride of tetramethyl thionine), thionine blue (hydro- chloride of trimethyl-ethyl-thionine). XL Azines. — (a) Eurhodines. — Example, neutral violet (hydrochloride of dimethyl-diamido-phenazine). {h) Safranines and indulines. — Examples, safranine, naphthalene red, mauve, induline. XII. Artificial Indigo. XIII. Quinoline Colours. — Examples, quinoline red and quinoline yellow, aldehyde green. XIV. Acridine Colours. — Examples, acridine yellow, acridine orange, phosphine. XV. Thiohenzevyl Colours. — Examples, thioflavine, primuline. To these must be added a number of colours like canarine and murexide, which either do not fall within the above divisions, or their constitution is insufficiently known. The group reagents used by Schultz and Julius are — (1) a solution con- taining 10 per cent, tannin and 10 per cent, sodic acetate ; (2) zinc dust and hydrochloric acid, or zinc dust and ammonia ; (3) 1 per cent, solution of § 64.] COLOURING-MATTERS. Ill chromic acid ; (4) 1 per cept. solution of chromic acid and 5 per cent, sulphuric acid.* Group I.-COLOURS SOLUBLE IN WATER. A. Pkecipitated by Tannin Solution : Basic colours. Reduce aqueous solution by zinc dust and hydrochloric acid, and put a drop or two of the decolorised solution on filter paper. If the colour does not quickly return on exposure to air, the spot is touched with a drop- of 1 per cent, chromic acid solution. (L) The original colour quickly reappears on exposure to air; azine, oxazine, thiazine, and acridine colours. RED.— TOLUYLENE RED, SAFRANINE, PYRONINE, ACRIDINE RED. ORANGE AND YELLOW. — PHOSPHINE, BENZOFLUORINE, ACRIDINE YELLOW, ACRIDINE ORANGE. GREEN.— AZINE GREEN. BLUE.— METHYLENE BLUE, THIONINE BLUE, TOLUIDINE BLUE, MELDOLAS BLUE, MUSCARINE, NEUTRAL BLUE, BASLE BLUE R and B B, NEW METHYLENE BLUE G G, NILE BLUE, CAPRI BLUE, FAST BLACK. INDAZINE M, METAPHENYLENE BLUE B, PARA- PHENYLENE BLUE, INDAMINES. VIOLET. — MAUVE, AMETHYST, NEUTRAL VIOLET, FAST NEUTRAL VIOLET, PRIME, PARAPHENYLENE VIOLET, INDAMINES. (2. ) The colour reappears very slowly, or not at all, on exposure to air, but returns on spotting with 1 per cent, chromic acid solution; triphenyl- methane colours and basic phthalenes. RED.— MAGENTA, ISORUBINE, RHODAMINE, ANISOLINE. GREEN.— MALACHITE GREEN, BRILLIANT GREEN, METHYL GREEN, IODINE GREEN. BLUE.— VICTORIA BLUE B, VICTORIA BLUE 4 R, NIGHT BLUE. VIOLET.-METHYL VIOLET, CRYSTAL VIOLET, HOFMANN'S VIOLET, BENZYL VIOLET, ETHYL PURPLE, REGINA PURPLE. (3.) Original colour does not return at all. YELLOW AND BROWN.-AURAMINE (this colour, when a solution is reduced, and a drop placed on filter paper and warmed over a Hame until dry, gives a beautiful violet) ; THIOFLAVINE T (this colour is reduced with difficulty) ; CHRYSOIDINE, BISMARCK BROWN. * Systematic Survey of the Organic Colouring- Matters, by Drs. G. Schultz. and P. Julius. Translated by Arthur G. Green, F.I.C., F.C.S. London, 1894. 112 foods: their composition and analysis, [§64. B. Not Precipitated by Tannin .Solution : Acid colours. The aqueous solution is reduced with ziuc dust and ammonia, or zinc dust and hydrochloric acid, and a drop of the decolorised solution put on filter paper. If the colour does not quickly return on exposure to air, the spot is touched with a drop of chromic acid solution (1 per cent. CrOs + 5 per cent. H2SO4), warmed over a flame, and then held in the vapour of ammonia. {a.) The solution is decolorised. (a.) The colour quickly reappears on exposure to air; SULPHONATED AZINES, OXAZINES, THIAZINES, &c., SOLUBLE IN- DULINES, SOLUBLE NIGROSINES, RESORCIN BLUE, AZURINE, THIOCARMINE, BASLE BLUE R S and BBS, GALLAMINE BLUE, GALLOCYANINE, GALLANILIC INDIGO P S, INDIGO CARMINE, SAFROSINE, AZOCAR- MINE, MIKADO ORANGE. (/3.) Tlie original colour does not reappear on exposure to air, or only very slowly, but returns with chromic acid and exposure to ammonia vapour The aqueous solution of the dyestuff is acidified and shaken with ether. (a.) Ether extracts more or less completely the phthaleins and aurine, such as URANINE, CHRYSOLINE, EOSINE, ERYTHRINE, PHLOXINE, ERYTHROSINE, ROSE BENGAL, CYCLA- MINE, AURINE, CORALLINE. (Of these, eosine, erythrine, and phloxine contain bromine ; erythrosine, rose bengal, and cyclamine contain iodine ; and the bromine or iodine may be set free by the action of strong sulphuric acid.) {h.) Ether does not extract tlie sulphonated triphenyl-methane colours — that is the following :— ACID MAGENTA, ACID VIOLETS, FORMYL VIOLETS, ALKALI BLUES, SOLUBLE BLUES, ALPINE BLUE, PATENT BLUE, FAST GREEN-BLUISH, ACID GREENS, GUINEA GREENS, CHROME VIOLET. {k.) The original colour does not appear at all; azo, nitro, nitroso, and hydrazine colours. Heated on platinum foil, nitro colours deflagrate with the production of coloured vapours ; such nitro colours are — PICRIC ACID, VICTORIA YELLOW, AURANTIA, MAR- TIUS YELLOW, NAPHTHOL YELLOW S, BRILLIANT YELLOW, AUROTINE. On the other hand, azo, nitroso, and hydrazine colours burn quietly or deflagrate slightly, giving off at the same time (like the nitro colours) coloured vapours. Small pieces of unmordanted cotton are soaked in the solution, and the resulting coloured cotton submitted to the action of warm soap. The substantive azo colours, such as ORIEL YELLOW, CLAYTON CLOTH RED, ALKALI BROWN, ATLAS RED, VIOLET BLACK, COTTON SCARLET, NAPHTHYLENE RED, HESSIAN PURPLE, CONGO- YELLOW, CONGO-RED, BENZO-PURPURINE, resist the action of warm soap, but the ordinary azo colours are stripped by them. § Gia.] COLOURING-MATTERS. 113 {X. ) Colour not decolorised by zinc and ammonia, but changed to brownish- red. Original colour returns quickly on exposure to air — ALIZ- ARINE S, ALIZARINE BLUE S, CCERULINE S. (ft. ) Colour by the action of zinc and ammonia slowly and incompletely disappears-CLAYTON YELLOW, THIAZOL YELLOW, TUR- MERINE, MIMOSA. <».) Colour not altered by zinc and ammonia ; very slowly or not at all by zinc and hydrochloric acid— QUINOLINE YELLOW, PRIMU- LINE, THIOFLAVINE, OXYPHENINE. Gkoup IL-DYESTUFFS INSOLUBLE IN WATER. The colouring matter is treated with water, and a few drops of 5 per cent, caustic soda solution. (1.) The colour dissolves. The alkaline solution is heated with zinc dust and ammonia, and a drop is jjut on filter paper. The colour fades or is changed to light brown. (a.) Original colour reappears quickly on exposure to air — COERULINE, GALLEINE, GALLOCYANINE, GALLANILIC VIOLET B S, GALLANILIC BLUE P, GALLOFLAVINE, ALIZARINE BLUE, ALIZARINE BLACK, ALIZARINE, CYANINE, ALIZARINE-CYANINE BLACK, RUFIGALLOL. (6.) Original colour does not reappear on exposure to air— ALIZARINE, ANTHRAPURPURINE, FLAVOPURPURINE, ALIZARINE ORANGE, ALIZARINE BROWN, ALIZARINE BORDEAUX, ALIZARINE YELLOW G G & R, CHRYSANINE, SUDAN BROWN, PATENT FUSTIN, MYRTLE or RUSSIAN GREEN, GAMBINE R and Y, DIOXINE. (2.) The colour remains insoluble. (A) Soluble in 70 per cent, alcohol. {a.) Solution not fluorescent. On adding caustic soda (33 per cent.) to the alcoholic solution — IN DU LINE OPAL, NIGROSINE OPAL, ROSANILINE BLUE OPAL, and DIPHENYLAMINE BLUE OPAL change to reddish-brown; while INDOPHENOL, SUDAN II and III, and CARMINAPHTH are not altered. {13.) Solution fluorescent. On adding caustic soda as above, Magdala red fluorescence is destroyed ; while the fluorescence of the spirit eosine and cyanosine is unaltered. (B) Insoluble in 70 per cent, alcohol. INDIGO, ANILINE BLACK, PRIMULINE BASE. § 64a. Otto Witt and E. Weingartner classify aniline colours as follows : — I. Soluble in water. (a.) Basic colours. [b.) Acid colours, II. Insoluble in water. 114 FOODS : THEIR COMPOSITION AND ANALYSIS. § Gib. To isolate the basic soluble colours, the aqueous solution is alkalised by- baryta, soda, or x^otash, and the colouring-matter extracted by shaking out with acetic ether. The acid soluble colours are tested for, as follows : — To the aqueous solution is added an excess of calcined magnesia, and then a solution of mercuric acetate (20 per cent, strength), the whole boiled and filtered ; the filtrate is either coloured or colourless, should the filtrate be colourless, if an acid aniline colour be present, acidification with acetic acid will repro- duce the colour. The aniline dyes insoluble in water are dissolved either in soda solution or in alcohol. The chief basic colours in presence of baryta w^ater dissolve in acetic ether ; they arc equally soluble in ammoniacal amylic alcohol, which they for the most part colour ; but if not, the colour is reproduced by adding to the amyl solution acetic acid. They are also precipitable by tannin. These remarks apply to the following : — RED COLOURS.— FUCHSINE, TOLUYLENE RED, SAFRANiNE. YELLOW AND ORANGE COLOURS— PHOSPHINE, FLAVANI- LINE, AURAMINE, CHRYSOIDINE, VESUVINE. GREENS.-VICTORIA GREEN, MALACHITE GREEN, METHYL GREEN, METHYLENE GREEN. BLUES. -METHYLENE BLUE, VICTORIA BLUE. VIOLETS. — METHYL VIOLET, HOFFMANN'S VIOLET, MAU- VEINE. The acid aniline colours are not precipitated by tannin ; they are insoluble or but slightly soluble in acetic ether in the presence of baryta water, and in any case the addition of acetic acid does not increase the tint. They are also insoluble or but little soluble in amylic alcohol— the chief of these colours are REDS. — Eosine, cosine scai-let, phloxine, Bengal red, Biebrich's scarlet, croceine, Congo-red, coccinine, f)onceau from xylidiue, ponceau R, 2R, 3R, roccelline (R solid), Bordeaux B, acid fuchsine, coralline, aurine, benzo-purpurine (B) 4B, delta-purpui-ine SB, Congo-corinth, Congo-corinth B, erythrosine, ponceau S, purple N, heliotrope, rosazurine G and B, induline, nigrosine. YELLOW AND OP^ANGE. — Fluoresceine, benzil fluoresceine, picric acid, JNIartius yellow, naphthol S yellow, aurantia, chrysamine G and R (this colour is not very soluble in cold water), chrysophenine, tropffioline 0, tropseoline 0, methj'l orange (tropteoliue D), orange I, orange II, orange III (ethyl orange), Poirier's yellow, metaniline yellow, luteoline, citronine, orange (D Pi, P 3,229), chrysoine, man- darine, tartrazine, alizarine S. VIOLETS. — Acid violet, nitro-violet, heliotrope, induline, and nigrosine. BLUE. — Alkaline blue, soluble blue, benzo-azurine G and R, nitro-blue, alizarine blue S, indigo-carmine, soluble induUnes. GREENS.— Helvetia green. § 646. Stein's method of procedure to detect colours generally is as follows : — § 646.] COLOURING-MATTERS. 115 RED COLOURS. A. Heat with Ammonium Sulphide.— A greenish or bluish colour, which, by the action of baryta water, is changed into green — ALOES PURPLE. If the liquid becomes purple, ARCHIL is also present. B. Boil with a Solution of Aluminium Sulphate. a. The liquid is coloured red, with a beetle-green reflection— MADDER. (Confirm by spectroscope.) b. The liquid becomes red, but there is no reflection. Add an equal volume of sodium sulphite. (I.) It. /"s bleached. Presence of BRAZIL-WOOD, SANTAL, MAGENTA, CORALLINE SAFRANIN. C. Boil with Alcohol of 80 per cent. a. The liquid colours distinctly. If bluish -red, MAGENTA; if yellowish- red, SANTAL! N. b. Liquid colours very little, or not at all— BRAZIL-WOOD, CORAL- LINE, SAFFLOWER. (1.) Heat with lime water. Xo colour— SAFFLOWER. Red colour— BRAZIL-WOOD, CORALLINE. (2.) Heat with dilute sulphuric acid. Orange colour— BRAZIL-WOOD. Yellow turning to grey on addition of copper chloride — CORAL- LINE. (Confirm by spectroscope.) (2.) It is not bleached. COCHINEAL LAC-DYE, LAC-DYE, KERMES, ARCHIL. (a.) Boil with alcohol; liquid becomes red — ARCHIL. If it only faintly colours, or at least if tlie colour is not decided, it may be COCHINEAL, LAC-DYE, KERMES. lb.) Heat with baryta water; no change — LAC-DYE ; the liquid becomes red-COCHINEAL, KERMES. (c ) Heat with lime water: a red colour— KERMES ; a violet colour— COCHINEAL. YELLOWS. Heat with a dilute solution of neutral ferric chloride. iV'.5.— This test must be applied when the colouring-matter is separated in a fairly pure state. A. Colour but little altered— ANN ATTO, TURMERIC, ANILINE YELLOW, PICRIC ACID, NAPHTHALINE YELLOW. Test with a drop of concentrated suljihuric acid: a blue or green colour is produced— ANN ATTO. If the .'^pot becomes at once, or after a little time, more or less brown or red, then add alcohol, with a few drops of hydrochloric acid and some boric acid. (a.) Liquid becomes of an intense pink colour, and on diluting with water, there is a reddish-yeUow colour — TURMERIC. (6.) Pale pink colour, on dilution with water, crimson — ANILINE YELLOW. 116 foods: their composition and analysis. [§646. (c.) There is no change of colour on the addition of hydrochloric acid, &c. Heat with amuioniacal copper solution, blimh-green; confirm the pre- sence of picric acid by the cyanide of potassium test, a blood-red colour— PICRIC ACID; the colour becomes an olive green— NAPHTHALINE YELLOW. B Various shades of colour from green to almost black— MADDER YEL- LOW, FUSTIC, FUSTET, QUERCITRON, FLAVIN, BER- RIES, WELD. Boil with aluminium sulphate, with the addition of an equal volume of water ; liquid becomes yellov/, with a red reflec- tion— MADDER YELLOW, with tin. Yellow, with a bluish-greea reflection— FUSTIC. Liquid yellow without reflection. Heat with baryta water, a red colour— FUSTET. The colour is only darkened. Boil with glacial acetic acid. On cooling, if the liquid is yellow, or greenish-yellow— ENGLISH FLAVIN. ' If the solution is not at all, or only faintly, coloured, boil with basic lead acetate. This, with regard to fabrics, if a tissue is di/ed with WELD, the tissue will not change colour; if with QUERCITRON or BERRIES, the tissue will change to orange-brown; but articles of food will be scarcely coloured with these substances. GREEN COLOURS. If the green colouring-matter is not soluble in water, it is probably chlorophyll, unless indeed it is a mineral colouring substance. Chlorophyll best recognised by the spectroscopic characters of its alcoholic solution (see p. 104). If not chlorophyll nor a mineral substance, then A. Boil with a moderately concentrated solution of potassic cyanide. (a.) Colour changes into brown or yellow— ANILINE GREEN, GREEN, containing INDIGO-SULPHURIC ACID (carmine green). (&.) Does not change, or changes into a brownish- or yellowish-ffreen — GREEN, containing INDIGO with or without CARMINE GREEN. B. In any of the foregoing cases add an equal volume of water, and then a solution of aluminium sulphate until an abundant precipitate is form ed ; filter and wash . (Excess of the precipitate must be avoided. ) (a.) Filtrate yellow or reddish— ANILINE GREEN. (b.) Filtrate blue— (I) precipitate colourless— CARMINE GREEN with PICRIC ACID; (2) precipitate yellow — CARMINE GREEN with a vegetable yellow. Dissolve the yellow precipitate in water, add sulphuric acid, and filter : a green fluorescence — FUSTIC; no fluorescence — WELD, TURMERIC. Test for turmeric in the original substance by boiling with alcohol, and adding boric and hydrofluoric acids. (a.) Filtrate colourless, precipitate yellow — INDIGO. (b.) Filtrate blue; if the precipitate is colourless, PICRIC ACID may be present, and should be tested for; if coloured, there is probably a vegetable colour present. TABLE I.— Effect of Volatile Solvents Acting on Acid Solutions. Name. The Add Watery Fluid ia In Water aiidifled with Snlphurio Acid dissolved or suspended, Is extracted by Petrolenm Etber. Benzine. Ether. Chloroform. Amy) alcohol Aniline red. Reddish. Nothing. Only impurities. Traces which, when dissolved, are colour- less, but, on evapora- tion, red. As ether. Is coloured red, extracts easily, gives a residue. Aniline violet. But little coloured, for the greater part sepa- rates. Nothing. Only impurities. Is coloured lilac, and leaves behind traces of a violet residue. But 1 i ttle coloured ;gives but small residue. Extractsboth dissolved and sus- pended, is colon red violet, and gives a residue. Aniline new violet. Violet from the solution of a portion. Nothing. Only traces. As benzine. As benzine. As with aniline violet. Aniline blue (insoluble). All separated; water colourless. Nothing. Traces which, on ex- posure to air, are coloured, and give a blue residue. Extracts copiously; is coloured strongly pale gold. As ether. As ether, but a larger quantity dissolved. Aniline blue (soluble). Blue, Nothing. As with the insoluble blue, but residue greenish-blue. Only traces. Only traces. As with the insoluble. Aniline yellow. Bright yellow. Is coloured yellow; de- posits yellow crystals spontaneously and on evaporation. As with petrolenm ether. As with petroleum ether, only in greater quantity. Picric acid. Bright yellow. Is not coloured, but leaves on evaporation a yellow residue. As with petroleum ether. Is coloured yellow, and gives a yellow residue. As with petroleum ether. As with ether, only in greater quantity. Styphnic acid. Bright yellow. As with picric acid, but a smaller portion is dissolved. As with petroleum ether. ? As with petroleum ether. Is dissolved easier than picric acid, the solution yellow. Chrysammic acid. Yellow neutral, solu- tion red. Nothing. Is coloured yellow; the separated benzine is red when shaken with caustic potash. ? As with benzine. As with benzine, only easier soluble. Aniline orange. Bright yellow, with greenish flocks. Only impurities. Is coloured yellow, and gives a brownish- yellow residue. As benzine, but easier soluble. As benzine. Dissolves easily; solution greenish - yellow, residue brownish. Havanna brown. Dark brown. Only impurities. Is colon red blue-yellow ; residue amorphous brown. Extracts very little. Extracts very little. Dissolves easily; solution deep red-brown, residue brown amorphous. Vesuvin. Brown. Only impurities. Is coloured yellow; re- sidue brown amor- phous. Takes less up than benzine. Takes less up than benzine. Dissolves easily; solution deep red -brown, residue amor- phous. Coralline. Yellow, but little dissolved. Only impurities. Only impurities. Dissolves much; resi- due orange. Solution yellow to deep brown. As ether and chloroform. TABLE II.— Effect of Volatile Solvents Acting on an Alkaline Solution. Name. Th8 Ammontacal SoInUon is coloored Dissolved in Ammoniacal Water, and ejtracted by Petroienm Ether. Benzine. Ether. Chloroform. Arnyl Alcohol. Aniline red. Almost colourless. Becomes fluorescent, but only dissolves a trace, probably im- purities. Is coloured yellow, strongly fluorescent; residue red. Is coloured bluish; the residue as with benzine. Is coloured deep red; the resi- due blue-red. Aniline violet. Almost colourless. As with aniline red. As with aniline red, only the residue is Is coloured bluish; the residue violet. Is coloured a violet- blue; residue Ukewise violet. Is coloured a deep violet-red; residue violet. Aniline new violet. Almost colourless. Nothing. Nothing. Only traces. Less than from acid; the resi- due the same as from acid. Aniline blue (insoluble). Discoloured. Is coloured red -brown; residue bluish. As with petroleum ether. As with petroleum ether. More is dissolved, but other- wise as petroleum ether. Aniline blue (soluble). Reddish. Nothing. Only a little extracted. Only a little extracted. Is coloured yellow, and gives a blue residue. Aniline yellow. Dark brown. At first yellow, but again separating, and becoming colourless. Extracts traces. Is coloured yellow, but little is extracted. As with henzine. Takes up less than out of acid; is coloured yellow, residue yellow. Aniline orange. Brown. Nothing. Extracts traces. Is coloured yellow, and leaves a small yellow residue. As with benzine. Extracts less than from an acid solution. Havanna brown. Clear brown and dis- coloured. Is fluorescent in the green; residue brown- ish. As with petroleum ether. Extracts less than with benzine. Becomes deep brown, with a green fluorescence; residue Vesuviu. Clear brown. Is coloured yellow; residue brown. Is coloured orange; residue brown. Is coloured yellow; residue less, hut as with benzine. Is coloured deep brown; resi- due brown. Coralline. Splendid purple colour. Only impurities. Only impurities. Is coloured pale yellow; residue red-brown, is coloured purple by NH3. Extracts less than out of an acid fluid. Is coloured raspberry- red; re- sidue as from ether. TABLE III.— Various ColoTir Beactions. Nius. Concentrated Sulphario Acid. Offlcinal Nitric Acid. DiBBOlved in Snlphnrio Acid Fluid. TannloAcld. In Hydrochlorio Acid Solation. Iodine. Iodide of Pota^aium and Iodide of Mercury and Potassium.' Hatin Chloride. Gold Chloride. Aniline red. Dissolves yellow. Dissolves • green, then brown, on evaporation again red. Dissolves red-violet, is quickly discoloured. Greenish coloration. No precipitate, or but a alight one. Violet-red precipitate. No turbidity. No precipitate. Greenish precipitate. Aniline violet Dissolves dark yellow to brown. Dissolves brown, then deep green, on evapora- tion red-brown. Dissolves violet, later almost colourless. Greenish coloration. Blackish precipitate. Blue precipitate. Black precipitate. Aniline new violet. Dissolves blocd-red. DUsolves deep blue.. Dissolves violet. Greenish coloration. No precipitate. No precipitate. No turbidity. No turbidity. Aniline blue (in- solnble). la coloured blood-red to brown. It is dissolved blue. Is dissolved without coloration. Greenish coloration. No precipitate. Aniline blue (soluble). Is coloured blood-red to brown. Dissolved blue. Is dissolved without coloration. Brown coloration. No precipitate. No precipitate. Aniline yellow. Dissolves yeUow. Dissolves yellowish. Dissolves orange. No precipitate. No precipitate. Slight turbidity. Slight turbidity. Picric acid. Dissolves brownish-red. Dissolves yellow. Dissolves yellow. No precipitate. No precipitate. No precipitate. No turbidity. Styphnic acid. Dissolvesalmost colourless. Dissolves colourless. Dissolves yeUow. No precipitate. No precipitate. No precipitate. Chrysammic acid. Dissolves with separation of a violet powder. Dissolves greenish-yellow. Dissolves red. No precipitate. No precipitate. Dissolves brownish. Dissolves pale yellow. Dissolves reddish. No precipitate. No precipitate. SUght turbidity. SUght turbidity. HavamiabHwn. la not made darker. Dissolves brown. Dissolves with difficulty brownish. No precipitate, but a red colour. Brown colour when shaken with ammoniaoal solution. Red-brown precipitate. Brown precipitate. Brown precipitate. Brown precipitate. Vesuvin. Dissolves brown. Dissolves blood-red, later red-brown. Dissolves easily— colour, brown. No pecipitate, but a red Brown colour when shaken with ammoniacal solution. Brown precipitate. Brown precipitate. Coralline. Dissolves yellow. Dissolves yellow. Dissolves purple. No precipitate. No precipitate. No turbidity. Slight turbidity. No turbidity. No turbidity. §65.] Asn. 117 BLUE COLOURS. Dissolve out the colouring-matter with alcohol of 80 per cent., or treat the substance itself with alcohol, allowing it to remain in the liquid, and add a few drops of hydrochloric acid: a red colour— LOGWOOD. Confirm by adding to an alcoholic solution ammonia and then alum ; a blue or violet precipitate should result. If the liquid does not dis- solve any of the colour, it is most probably Prussian blue, but it may also be indigo. If, on the other hand, although some of the colour is extracted, vet the substance still remains blue, it is probably ANI- LINE BLUE or INDIGO-SULPHURIC ACID. Add strong sul- phuric acid, with INDIGO-SULPHURIC ACID, no change. The colour changes into a yellower reddish-brown in presence of ANI- LINE BLUE. On heating with sodium carbonate again, INDIGO is not changed, but PRUSSIAN BLUE is changed into yellow or brown. VIOLET AND PUEPLE COLOURS. Heat with ammonium sulphide. A. The tissue is bleached; soluble ANILINE VIOLET, MAGENTA, with INDIGO CARMINE. These two may be distinguished by the action of boiling alcohol: SOLUBLE VIOLET I'emains violet, MAGENTA becomes red, the substance becomes brownish-red. Presence of MAUVE, or HOFFMAN'S VIOLET. These may be distin- guished by the addition of hydrochloric acid, which colours HOFF- MANN'S VIOLET yellow, but MAUVE becomes purple. B. Turns olive brown — ALKANET. (Confirm by spectrum.) D. Hardly any change. Presence of ARCHIL, ARCHIL with INDIGO, LOGWOOD, or MADDER. Boil with alcohol— a. The solution is pink, and changes to violet on the addition of ammonia ARCHIL; if the ARCHIL is accompanied with INDIGO, hot chloroform will colour blue. b. Solution in alcohol remains colourless. Heat with dilute hydrochloric acid: LOGWOOD is coloured red, and may be further identified by test already given; if INDIGO is associated with it, hot chloroform will colour blue. With hydrochloric acid, madder, if it is coloured at all, becomes yellow. The accompanying Tables will also be found useful in the identification of orcranic colours. THE MINERAL MATTERS OR "ASH" OF FOOD. ANALYSIS OF THE ASH OF ORGANIC SUBSTANCES. § 65. As a general rule, testing the ash for abnormal metals and alkaline earths is necessary, and more especially if the ash present any unusual character, whether in weight, colour, or 118 foods: their composition and analysis, [§65. solubility. Leaving for tlie present the s;)ecmZ tests, the number and nature of the constituents which require to be determined for the purpose of the food-analyst, vary according to the par- ticular substance under examination, e. g. — In all substances, the percentage. In such fluids as milk, the alkaline phosphates and the chlorides. In seeds, such as wheat, cocoa, &c., the ioial phosphoric acid. In beer-ash, the amount of common salt. In bread-ash, the presence or absence of alicmiiia, 7nagnesia, and proportion of silica to alumina. In tea-ash, the alkalinity, the iron, the silica, and proportion of soluble to insoluble ash. In coffee-ash, likewise the proportion of soluble to insoluble ash, but the presence or absence of silica becomes also of im- portance. From these illustrations (which might be multiplied) it follows that, for the purposes of the food-analyst, the general constitution of the ash will be sufficiently known when the following deter- minations have been made : — (1.) The total percentage of ash. (2.) The total percentage of ash soluble in water. (3.) The total percentage of ash soluble in acid. (4.) The alkalinity of the ash. (5.) The percentage of chlorine. (6.) The percentage of phosphoric acid. (1.) The Total Percentage of Ash. — Of the various methods of estimating an ash, the simplest and most practical appears to be — to place a sufficient quantity of the substance to be burnt in a capacious platinum dish, and to consume at the lowest possible temperature by heating with a ring burner.* The quantity to be taken is regulated by the amount of ash in the substances. For example, flour, containing only -7 per cent, of ash, would give with 50 grms. -35 ash, which is about as small a quantity as it is possible to work with conveniently, whilst in the case of coffee, tea, and cocoa, from 5 to 20 grms. is for most purposes ample. ■'' If the sulphuric acid in the ash is not to be determined, a wide glass tulie (such as the chimney of a common paraffin lamp) adjusted over the dish, by its ]>owerful up-drau^ht greatly expedites the operation ; but if the sulphuric acid is to he determined, the impure gas of commerce renders the results too high. It is, however, of course open to the analyst to make the gas pass through a proper absorption-apparatus, or to use as a fuel, alcohol. § 66.] Asn. 119 (2.) The Soluble Ash. — The asli is boiled up two or three times with water in the same platinum dish ; filtered, and the filtrate evaporated to dryness, heated to dull redness, and weighed. (3.) The Ash Soluble in Acid. — The portion of ash insoluble in water is boiled up with HCl, and filtered from the sand; the latter is washed, dried, and weighed. (4.) The Alkalinity of the Ash. — The solution in water from (2) is coloured with cochineal, and titrated with d. n. acid : the result may be usually expressed as potash. (5.) The Percentage of Chlorine.- — The determination of chlorine in the ash usually gives results too low, especially if the sub- stance burnt is one, like bread, of difficult combustion, or con- taining substances which decompose chlorides at a red heat. Notwithstanding this defect, in a series of ashes burnt under similar circumstances, the amount of chlorine found gives fair comparative results. Should there be any special necessity for an accurate determination of chlorine, no volatilisation will occur in the combustion of most articles of food, if they are sim})ly well •carbonised and not burnt to a complete ash, and if the charcoal be finely powdered and extracted with plenty of boiling water. The chlorine may be determined gravimetrically by nitrate of silver, or more conveniently by a standard solution of nitrate of silver, using as an indicator neutral chrouiate of potash. Should alkaline phosphates be present, they must be first removed by baryta water. (6.) The Phosphoric Acid. — The usual method of determining phosphoric acid is to dissolve the ash in hydrochloric acid, evaporate to dryness, remove the silica, mix the acid filtrate with ammonia in excess, redissolve the precipitated earthy phosphates by acetic acid, filter off and estimate the insoluble phosphate of iron (and alumina, if present), precipitate the lime with oxalate of ammonia, and then in the fluid (free from lime and iron) precipitate the phosphoric acid, by the addition of ammonia and magnesia mixture. ^ (jQ. General Method of Determining all the Constituents of an Ash. — The best method of determining all the constituents of an ordinary ash is perhaps as follows : — A sufficient quantity of the ash (from 5 to 10 grms.) is placed in a flask, about 25 cc. of water added, and saturated with CO., ; the liquid is now evaporated to dryness, heated with a small quantity of water to dissolve the alkaline salts — the solution is filtered througli a small weighed filter, the filtrate evapoi-ated to dryness, the saline residue treated with a small quantity of water, and the calcium sulphate which separates out filtered through a weighed filter, and estimated ; the filtrate from this is put in a tared flask. 120 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 6G. made np to any convenient weight, and divided into five portions hy tveight, viz. : — (1.) For COg. — This is most accurately determined by the use of the following little apparatus : — The solution is. placed in a flask A, and suffi- cient acid is put into the short test-tube T, to more than neuti-alise the carbonate. A stout glass rod is passed through the doubly perforated caoutchouc cork, and supports the little tube in position. The carbonate solution is now boiled until steam hisses out of the tube G (which it is convenient to furnish with a G is then placed under the mouth of a graduated measuring tube filled with mercury, and it is at once seen whether all the air is expelled. The flame is with- drawn for a second, and the glass rod, which moves quite air- tight, is pulled a little up, so as to allow the acid tube to fall down and empty its contents into the alkaline fluid. The flame is again placed under the flask, and the COg boiled ou^, into the measuring apparatus, and measured in the ordinary way. Those who are not provided with gas apparatus will find it convenient to jacket their eudiometer with a tube open at both ends. The lower end is closed by the mercury in the bath ; the upper is placed under a water-tap, and a syphon is adjusted so as to prevent overflow. In this way the gas is rapidly cooled, and the whole determination, from first to last, need not take more than a quarter of an hour.f Instead of boiling the solution in this way, those who possess the mercury pump described and figured at page 70 (fig. 6), will find it more convenient to make the flask vacuous, then upset the acid, and collect the gases expelled. Fig. 19. Bunsen's valve*). * Bun sen's valve is made as follows : Take a piece of rather thick-walled india-rubber tubing, say, three inches in length ; work it, by the aid of a little spirit, on to any wooden rod which is of sufficient size to stretch it well ; then with a sharp chisel, by a single blow, cut a longitudinal slit ; if well made, it allows air to go one way with the greatest ease, but etFec- tually prevents a return. + Much ingenuity has been expended on the estimation of carbon dioxide, and the old method of estimating in light glass apparatus, by the loss of weight, is quite forsaken. The method given in the former edition of this work was to absorb the COj in a clear solution of ammoniacal calcium chloridcj collect the precipitated calcium carbonate, and titrate it with d. n. acid. § GG.] ASH. 121 (2.) For the sulphuric acid, determined by chloride of barium. (3.) For the phosphoric acid, determined as magnesian pyro- phosphate. (4.) For the chlorine, by precipitation as silver chloride. (5.) For the alkalies, by boiling in a platinvim dish with slight excess of baryta water, filtering, getting rid of the excess of baryta by ammonia and ammonium carbonate, evaporating the filtrate to dryness, converting the alkalies into chloiides, and determining their relative proportion from their total weight and their content in chlorine. This completes the analysis of the soluble portion of the ash. The insoluble Avill contain lime, magnesia, ferric oxide, alumina if present, silica, phosphoric, sulphuric, and carbonic acids. The main portion of the insoluble ash is dissolved in nitric acid, freed from silica in the usual way, evaporated again to dry- ness in a porcelain basin, dilute nitric acid added until the bases are completely dissolved, and strong fuming nitric acid added, until the solution begins to be turbid from the separation of calcic nitrate. Tlie turbidity is now destroyed by a few drops of dilute nitric acid, the solution warmed, and tinfoil added in small por- tions at a time, in weight about equal to the amount of ash taken. When the tin is fully oxidised, the solution is evaporated nearli/ to dryness, water is added, and the solution filtered ; the phos- phoric acid is retained in the precipitate — the bases are all in the filtrate. The precipitate is dissolved in strong potash solution, acidified with sulphuric acid, and freed from tin by hydric sulphide, concentrated to a small bulk, filtered if any further sulphide of tin separates, and the phosphoric acid deter- mined by magnesia mixture and ammonia.* The filtrate from the tin phosphate must be freed from lead (if the tin originally contained lead) by hydric sulphide, con- centrated, the iron and alumina separated and determined by ammonia, the manganese sepai^ated as binoxide by bromine- water, the lime by oxalate of ammonia as oxalate, and the mag- nesia determined in the usual way as pyrophosphate. A weighed portion of the insoluble ash must also be taken for the carbon dioxide, sulphuric acid, and sand. The carbon dioxide is determined in the manner already described. using as an indicator cochineal. Another method which has been proposed, is to make a combustion of the substance with potassic bichromate, and absorb the COo in the usual way in potash bulbs. A third, is to let the acid drop from a separating funnel on to the carbonates, and absorb the C'Oa as in the Last, an aspirator and drying tubes being used. (Annalen der Chemie, c\xx\i. 136-144.) All of these methods, in the author's opinion, are inferior to that given above in the text. * Thorpe's "Quantitative Analysis." Lond.,1877. 122 FOOaS : THEIR COMPOSITION AND ANALYSIS. [§ 66. The process just given is not quite accurate with regard to the estimation of the alkalies ; for Bunge * has shown that since the alkalies form insoluble compounds with the alkaline earths, a watery extract of the ash gives low results. For example, Bunge incinerated 300 cc. of cow's milk ; from a watery extract of the ash he obtained K2O -5436, Na^O -0700 ; while from a subsequent nitric acid extract of the same ash, KgO -0937, NagO 0-1162. If chlorides of the alkalies be heated with tribasic phosphate of lime, the soda is specially likely to combine with the lime in insoluble combination — in far less proportion the potash. Bunge recommends the following method : — The watexy ex- tract is decomposed with bai'yta water until a film forms on the surface of the solution, the mixture is warmed and filtered hot. The excess of baryta is got rid of by CO2, subsequent warming, and filtration ; the filtrate is evaporated in a platinum dish, the residue gently ignited, dissolved in a little water, filtered through a small filter, and evaporated with HCl in a small platinum dish. The chlorides are then ignited, weighed, and separated by platinum chloride. The hydrochloric or nitric sokition of the insoluble portion of the ash is evaporated to di-yness in a platinum dish, the residue again dissolved in a little of the acid and water, treated like the former Avith baryta water, and filtered hot. Ammonia and car- bonate of ammonia are now added, the liquid filtered, and the filtrate evaj)orated in a platinum dish, and ignited at the lowest possible temperature. The residue still containing a trace of alkaline earth, is extracted with water, evaporated with oxalic acid, ignited again, taken up with water, filtered, evaporated in a small platinum dish, ignited again, dissolved in a little Avater, and lastly, evaporated with HCl, and the alkaline salts separated by bichloride of platinum. Since a determination of the ash only gives those mineral sub- stances which are fixed in the fire, and destroys nitrates, and changes oxalates, citrates, and tartrates into carbonates, while other constituents, under the influence of heat, undergo a new arrangement, it becomes a question whether the ingenious method recommended by E. Laugiert in the analysis of sugar, would not be applicable in several cases. * Liebig's Annalen der Chimie u. Pharmacie, April 15, 1874. t Compt. rend., Ixxxvii. lOSS-1090. § 66a.] ASH. 123 M. Langier takes two portions of sugar, one for the ash, the other for the organic acids, the latter being exactly double the quantity of the former. To the larger quantity of the sample, dilute sulphuric acid is added drop by drop to set free the or- ganic acids ; the acidified sugar is mixed with pumice stone and exhausted with ether ; half of the ethereal solution is added to the ash obtained from the smaller portion, and evaporated down upon it and weighed. By this means M. Laugier thinks that he reconstructs the original salts in the sugar. This, however, cannot be entirely true. The other half of the ether solution is titrated with an alkali. METHODS OF ESTIMATING NITROGEN AND NITROGENOUS SUBSTANCES IN EOODS. § 66a. A complete analysis of foods, especially with the view of ascertaining their dietetic value, necessitates the following determinations : — 1. Total nitrogen. 2. Nitrogen as albumen. 3. Nitrogen as acid-amines. 4. Nitrogen as amido acids. 5. Nitrogen as nitrates and nitrites. Total Nitrogen. — The method of burning organic substances •with copper oxide and copper, so as to obtain all the nitrogen as a gas, and also the other well-known method, by which the substance is mixed with soda lime and burned, so as to decom- pose the nitrogenous substances into ammonia, are both too well known to be farther described. At the present time, nitrogen is most frequently determined by Kjeldahl's process. From 1 to 2 grms. of the substance are placed in a flask of hard glass, of about 500 cc. capacity, and moistened with 20 cc. of a mixture of 3 volumes of concentrated sulphuric acid and 2 volumes of Nordhausen acid, and a bead of quicksilver is added. The flask is closed with a glass marble, and heated on an asbestos millboard, or on a sandbath, until the mass boils ; it is kept gently boiling until a clear solution is effected ; this usually takes from one to several hours ; the flask is then allowed to cool, the contents are diluted with double their bulk of water, and alkalised with 80 cc. of soda lye, of sp. gr. 1-35 ; 25 cc. of a 4 per cent, solution of potassium sulphide, or so much of the latter as will precipitate all mercury in the form of sulphide, are also added, and, lastly, a little fine zinc powder. The liquid is now distilled, with special precautions, into a flask containing from 10 to 20 cc. of normal sulphtiric acid. The special precautions mainly consist in adapting a tube with a bulb to the condensing 124 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 66a, tube ; with such a bulb it is scarcely possible for any fixed alkali to be carried over mechanically. After about 100 cc. are distilled over, the normal sulphuric acid is titrated back, using, as ah indicator, cochineal with normal soda solution, diluted one-fourth. The difference found between the amount of alkali required to neutralise an equal bulk of normal sulphuric acid, before and after distillation, is reckoned as ammonia and cal- culated into nitrogen. Estimation of the Nitrogen from Albumen. — The method of Stutzer is probably the best. From 1 to 2 grms. of the finely divided substance are boiled for a few minutes with 100 cc. of water, or, if the substance contains much starch, it is simply digested on the water bath for from ten to fifteen minutes, then from 0-3 to 0-4 grm. of moist hydrate of copper is added, and, after cooling, the liquid is filtered through Swedish filter paper, the filter washed with water, and the residue with the filter submitted to Kjeldahl's process. The ammonia thus obtained is considered to be derived from albumen only, and the difter- ence between the total nitrogen and that from Stutzer's process gives non-albuminoid nitrogen, which is sufiicient for most purposes. Should the substance contain difficultly soluble alkaloids, it is first treated with absolute alcohol, acidified with acetic acid and heated to boiling, the alcohol decanted off" as far as possible, and the residue washed with a little warm alcohol ; then the substance is treated as before with water and copper hydrate. Estimation of Non-Albuminoid Nitrogen. — 10 grms. of the substance are boiled with about 300 cc. of weak alcohol (40 per cent.), acidified with a few drops of acetic acid, the flask being connected with an upright condenser. The boiling should continue for about one and a-half to two hours ; the liquid is then cooled, filtered, and made up to a definite bulk, of which aliquot parts are taken for the following determinations : — 1. Estimation of Ammonia. — One-third of the fluid extract is evaporated down to 50 cc, cooled, some recently calcined magnesia added, placed in a small flask attached to a tube loosely packed with glass wool, moistened with 10 cc. of normal hydrochloric acid, and the air evacuated by a mercury or a good water pump, and after a vacuum has been obtained, a clip is put on the india-rubber tube connecting the apparatus with the pump, and the apparatus put on one side for three days. By the end of that time it is to be presumed that any ammonia in the apparatus will have evaporated and have been condensed in the glass wool moistened with hydrochloric acid ; the glass wool tube is now disconnected, and the wool washed § 66a.] Asn. 125 with a little distilled water, the acid water is evaporated to dryness on the water bath, and the ammonium chloride deter- mined by the well-known method of precipitation with platinum chloride. 2. Estimation of Amido-acid Amide Nitrogen. — Another third of the extract is boiled with dilute hydrochloric acid (7 to 8 cc. of strong acid added to every 100 cc. of liquid extract) for from one and a-half to two hours. At the end of that time the liquid is cooled and treated with hypobromite of soda in a nitrometer, and the resulting gas measured. The difference between the nitrogen found by the first process and that by the second process equals the nitrogen as amido- acid amide. 3. Estimation of Amido-acid Nitrogen. — Another portion of the liquid extract is also boiled with hydrochloric acid, and then cooled and subjected in a suitable apparatus to the action of sodic nitrite. The resulting gas is submitted in a Hempel's burette to strong alkaline permanganate, which absorbs both NO and CO.,, and leaves, if the operation has been properly con- ducted, only nitrogen ; the gas is measured in the usual way. Since nitrite evolves nitrogen from both amido-acid amides and acid amides, the nitrogen of 2 must be subtracted from the nitrogen evolved in 3. By the processes described the following are estimated : — 1. Ammonia. 2. Nitrogen of amido-acid amides from difference between Nos. 1 and 2. 3. The nitrogen from amido-acids from the difference between Nos. 2 and 3. PART UK— CARBO-HYDRATES. STARCHY AND SACCHARINE SUBSTANCES. § 67. The carbo-hydrates are divided into three great groups, viz., that of grape sugar, that of the saccharoses, and that of cellulose. They mostly contain 6, or a multiple of 6, atoms of carbon; and hydrogen and oxygen in the proportion of 2 : 1 — that is, in the same proportion as the constituents of water. Recently, however, sugars have been synthetically formed con- taining 3, 4, 5, 7, 8, and 9 carbon atoms. I. The Grape Sugar Group. To this group belong sugars containing 3, 4, 5, 6, 7, 8, and 9 carbon atoms, known under the names of trioses, tetroses, pent- oses, hexoses, octoses, and nonoses. These possess the following general properties : — 1. They are all easily oxidisable, and reduce Fehling's solution. 2. Most of the glucoses under the action of dilute mineral acids form levulinic acid with humus substances. 3. Warming with acetic acid and phenyl-hydrazine they all form more or less difficultly soluble and crystalline " osazones." The formation of osazones takes place in two stages ; first, one molecule of the sugar unites by means of its aldehyde or ketone group with one molecule of phenyl-hydrazine to form the easily soluble hydrazone, thus : — Sugar. Phenyl-hydrazine. CHoOH[CH(OH)]3CH(OH)COH + H.N . NHlCcHs) Hydrazone. = CHoOH[CH(UH)]3CH{OH)CH N— NH(CcH5) + H2O with separation of one molecule of water. Then, in presence of an excess of phenyl-hydrazine, another molecule of phenyl-hy- drazine unites with the hydrazone with separation of a molecule of water and a molecule of hydrogen, thus: — Hydrazone. CH2(OH)[CH(OH)]3CH(OH)CH + CgHsNH— NH2 N-NHCCeHj) Glucosazone. = CHo(0H)[CH(0H)]3C— CH + H2 + H2O II II CcHsNH— N N— NHCgHs § 67.] SUGAR. 127 The osazones are distinguished from each other by their differ- ences of solubility and their melting-point ; glucosazone melts at 205°, lactosazone at 200\ galactosazone at 193°, maltosazone at 206°, arabinosazone at 160°, and sorbinazone at 164°. 4. The glucoses all form additive compounds with hydric cyanide, the cyanide thus formed being converted, by the action of saponifying agents, into acids, which acids on reduction yield aldehydes ; these are true sugars. 5. Several, not all, of the glucoses are capable of fermentation. An example of a triose is glycerose, C3Hg03 ; of a tetrose is erythvose, O^HgO^ ; of a pentose is arabinose, C^H^qO^ ; the hexoses are glucose, mannose, fructose, and galactose. 11. The Cane Sugar Group. To this group belong sugars of the formula C^^^ooOj^ — e.g.y cane sugar, milk sugar, maltose, and sugars of the formula ^'18^32016 (raffinose). III. The Carholiydrates. To this group belong the so-called polysaccharides, cellulose, starch, glycogen, gummy matters, and dextrin. Cane Sugar, C-^oHojOj^, occurs in a very large number of plants, but is only manufactured from beet-root, the sugar-cane, sorghum, and the sugar-maple. Its specific rotatory power is + 66*5. It crystallises from its solutions in water or dilute alcohol in anhydrous crystals, the specific gravity of which is 1-606. It is soluble in one-third of its weight of cold water, and is very soluble in hot ; in absolute alcohol it is insoluble, the solubility rising in proportion to the weakness of the alcohol. Thus, according to Scheibler, the numbers in Table IV. are the percentages of sugar dissolved, and the specific gravity of the solution at the common temperature of 14°. When solutions of sugar are boiled with dilute mineral acids, the sugar is split up into two glucoses : the one rotating to the right, hence named dextrose, the other rotating the plane of polarised light, to the left — levulose.* Long boiling with water has, to a slight degree, the same effect, and it is also shown in the action of ferments, when exposed to light. An uncorked solution of sugar (or one imperfectly sealed) will in a few days, according- to the temperature, show some degree of inversion. But a boiled solution of sugar, which, while actually boiling, has been her- * To this mixture of dextrose and levulose, the term " i,'ire?-< sugar " is applied, because the polarisation is the opposite of that of cane sugar ; for although glucose rotates to the right and levulose to the left, yet the latter is so much stronger that the solution polarises to the left. 128 FOODS : THEIR COMPOSITION AND ANALYSIS. [§67. r cent, alcohol. ^^'^ cc. of the s contain . 85 -S 5 . . . 82-4 10 . 79-4 15 . 76-5 20 . 73-4 25 . 69 8 30 . 66 35 . 61-6 40 . 56-7 45 . 51-6 50 . 45-7 55 . 39-6 60 . 32-9 65 . 25-6 70 . 17-8 75 . 11-2 80 . 6-4 85 . . 2-7 90 . 0-7 95 . 0-2 97-4 0-08 100 . OOO TABLE IV.— SOLUBILITY OF SUGAR IN ALCOHOL OF DIFFERENT STRENGTHS. Specific gravity of the saturated solution. 1-3248 1-2991 1-236 1-2293 1-1823 1-1294 l-OSO 0-9721 0-8931 0-8369 metically sealed, -will keep unchanged for years. Citric acid iu a 2 per cent, solution readily inverts cane sugar; it does not act on milk sugar, a fact utilised in the detection and determination of mixtures of these sugars.* Carbon dioxide, especially under pressure, inverts sugar, f Pure cane sugar, if free from glucose, undergoes no change of colour when boiled with the alkalies ; if, however, glucose be present, there is a very decided change. Sugar forms a few well established compounds with bases, and many with salts. The most definite of the sugar compounds combined with bases are those which it forms with baryta and lime. If a solution of sugar be boiled down with sulphide of barium or baryta, a sandy precipitate forms, having the com- position Cj^oHogOjjBaO ; and on decomposition of this with COg, pure sugar is obtained. A commercial process based upon this reaction is in use in order to recover the sugar from molasses, and it may be employed in certain cases in the laboratory with advantage. There is a monobasic lime sucrate (OjoHogOj^CaO) corre- sponding to the barium compound, and a tribasic sucrate of lime * Stokes and Bodmer, Analyst, x. 62-65. t V. Lippmann, Ding. Poly. J., ccxxxvij. 146-163. § 68.] SUGAR. 129 (OjoHgpOjiSCaO). Crystalline compounds are also easily obtained with certain sodium salts ; thus, there is a chloride of sodium compound CjoHo20^^]SraCI.2H20 ; and another having the formula 2CjgHo20j^3NaC1.4H20. An iodide of sodium compound may be obtained in lavge crystals having the following composition : 2C,oH220„3NaI.H20. ^ Sugar heated to 160° melts to a colourless liquid; on cooling, the melted mass is at first clear and transparent, but in a little time it becomes crystalline and opaque. At about 170° to 180° it loses water, and is said to be transformed into dextrose and levulosan ; as the heat is increased, water is continually being lost, and more or less brown products are formed (see " Caramel," p. 108). If sugar is fused with zinc chloride, a liquid is obtained which yields, on distillation, aldehyde, acetone, metacetone, formic acid, acetic acid, furfurol, and apparently mesityl oxide ; carbon dioxide, carbon oxide, and hydrocarbons are also formed ; and there is also a sublimation of crystals, hexmethylbenzene, Cg(CH3)g. [Lijjpmann.) Bromine, according to E. Reichardt, transforms one-tliird of cane sugar into gluconic acid, one-third into glucose, and the remainder into gum. § 68. Adulterations of Sugar. — Loaf sugar is, as a rule, chemi- cally pure. It is probably, indeed, the purest food-substance in commerce, and a large quantity may be burnt up without obtaining a trace of nitrogen, and without leaving any residue. The only sugars that may be impure are the " raw " sugars. The adulterations of sugar usually enumerated are : Glucose or starch sugar, sugar of milk, dextrin, chalk, plaster, sand, and various species of flour ; few of these have been found of late years. A new adulteration is the colouring of sugars with yellow and yellow-brown aniline dyes; this is somewhat common. The detection is on the principles detailed in the chapter on Colouring- matters. One of these dyed sugars is at once detected by moistening the sugar with hydrochloric acid; a red colour is immediately produced. To detect glucose (dextrose) in the presence of other sugars, B. Bottger* mixes the solution with an equal quantity of car- bonate of soda solution [1 of the salt to 3 of water], and then adds a little basic bismuth nitrate, boils, noticing whether there is any blackening, which is taken as an indication of dextrose. E. Bruckef has modified this method, so as to eliminate any blackening (which might occur from the sulphur in albuminous matters), by using potassium bismuth nitrate, which precipitates albumen. The reagent is made by dissolving basic bismuth * Journ. fiir praht. Chem., Ixx. 432. t Wien Akad. Ber., 1875, 52. 10 130 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 68. nitrate in a hot solution of potassic iodide with the addition of hydrochloric acid. The albumen precipitated by the re-agent is of course filtered off, and the filtrate is boiled. The simplest and best method, however, of detecting starch sugar when mixed mechanically with cane sugar, is undoubtedly that recommended by P. Casamajor.* The suspected sugar is thoroughly dried, and is then treated with methyl-alcohol which has been saturated with starch sugar. 100 cc. of methylic alcohol of 50° strength dissolves about 57 grms. of starch sugar, the lOQ cc. becoming in volume 133 cc. Such a saturated solution dissolves cane sugar readily enough, but leaves starch sugar undissolved. After stirring the sugar in the methylic alcohol for about two minutes, the residue is allowed to settle, and the clear solution decanted. The residue is now washed with the same solution, and after stirring and allowing the residue ta settle again, if starch sugar wei-e present there will remain a certain quantity of chalky white specks, accompanied by a fine deposit of starch sugar. By collecting this on a filter, and ■ ■washing rapidly with nearly absolute methyl, approximate quantitative results may be obtained. The best method of detecting dextrin when mixed with sugar, has been specially studied by Scheibler.t He took a sugar which gave the following numbers to analysis : — Per cent. Water, 3-35 Ash, 1-73 Organic roatter, 2 '32 Sugar, 92-60 and mixed this sugar with various pi'oportions of dextrin — from 1 to 3 per cent. — and examined the behaviour of the sugar, both optically and chemically. The polarisation indicated the follow- ino- amounts of cane sugar : — 0-0 0-5 10 20 3-0 per cent, dextrin 92-6 93-4 940 95-6 96-3 ,, sugar. Thus, a sugar adulterated with 3 per cent., if examined optically, would indicate 9G-3 per cent, of cane sugar instead of 92-6. On inverting the sugar, there were great discrepancies, quite enough to make even an inexperienced observer suspect something •wrong. Thus with the same amounts of dextrin, the pure sugar * Chemical New% ISSO. + " The Sugar Cane." 1871, p. 469. § 68.] SUGAR. 131 showing, before inversion, 9G-3 per cent, of sugar, and after inversion, 67 '0 : — 0-5 1-0 2-0 0-0 per cent, dextrine. Direct, 93 --t 94-0 95 6 96-3 ,, siis^ar. After inversion, 804 TS'O 731 670 ,, ,\ It was no use to experiment beyond 3 per cent., because it •was then impossible to clarify the solution by lead acetate su^fficiently for the purposes of optical analysis. Scheibler summarised his results as follows : Dextrin may be detected by its thus raising the degree of rotation, by the great difference of the results before and after inversion, by the blue colour it gives with iodine (altliough there are dextrins which, when added to sugar, may show this test imperfectly), by the impossibility of clarifying the liquid should any amount be present, by lead acetate, and lastly, by partial separation of the dextrin by animal charcoal. Insoluble mineral matters, such as sand, present in low-class sugars as an impurity, may be I'cadily detected by simple solution of the sugar and filtration. Gummy matters may also be separated by precipitation by alcohol in the way to be described in the article on "Tea;" mineral matters, generally, may be detected in the ash. Beet sugars, and to a less degree cane sugars, will contain a large amount of potassic and sodic carbon- ates, arising from the decomposition of the citrates, malates, oxalates, &c. Beet sugars may also contain nitrates. Cane sugar leaves an ash containing but little soda, with much more lime, magnesia, iron, and alumina. Thus, the following is the asli of raw cane and beet sugars, obtained in the following manner : All the mineral matters in the sulphuric acid residue of a large sugar factory were kept for a whole yeai', and analysed at the end of the year. Of course the carbonates, nitrates, and chlorides have all been decomposed, and the analysis is true only with regard to the bases. Cane Sugar Ash. Beet Sugar Ash. Per cent. Per cent. Potash 28-79 3419 Soda -87 11-12 Lime, 8-83 3-60 Magnesia, . . . .2 73 -16 Oxides of iron and ahimiuiiim, 6 90 "28 Sulphuric acid [anhyd.], . . 43-G5 48-85* Dividing each factor of the ash of the beet sugar by that of the cane, we get the following proportions for the bases : — * J. Wallace, Chem. News, xxxvij. 75. 132 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 69. Cane. Beet. Potash 1 1-19 Soda, 1 12-78 Lime, 1 "41 Magnesia, 1 '06 Oxides of iron and aluminium, . . 1 '04 In other words, the potash is almost equal in the two ashes, but there is nearly thirteen times more soda in beet ash than in cane sugar ash ; lime, magnesia, and oxides of iron and aluminium, are in very small quantities in beet sugar ash. An analysis of the ash of a Demerara cane sugar growing near the sea-coast, by Di*. Wallace, is as follows : — Per cent. Potash, 29-10 Soda, 1-94 Lime, 15-10 Magnesia, 3-76 Sulphuric anhydride, 23-75 Phos])horic acid, 5-59 Chlorine, 4-15 Carbon dioxide, ...... 4-06 Iron peroxide, ...... '55 Alumina, ....... "65 Sihca, 12-38 101-03 Deduct oxygen = chlorine, ... "93 10010 If sugar be ever adulterated by any of the starches, so clumsy a fraud is readily detected by a microscopical examination, and the use of iodine to the residue obtained by dissolving the sugar in cold water, and then filtering. § 69. Full Analysis of Sugar. — The full analysis of a raw sugar consists in : — 1. Determination of the water driven off at a heat not exceed- ing 55° to 60°. 2. An optical estimation before and after inversion. 3. Titration with copper oxide before and after inversion. 4. Estimation of the organic acids by treating with sulphuric acid, and shaking up this acid extract in the tube figui'ed at page G9 with ether, until it has dissolved out all the organic acids. 5. Titration of the organic acids with d. n. soda or potash. 6. Estimation of any insoluble matter, whether organic or inorganic. 7. Estimation of the ash and its constituents. 69.] 133 It may also be necessary to estimate tlie matters precipitated by basic lead acetate ; it would be, however, quite sufficient for commercial, and, indeed, for most purposes, to merely estimate the percentage of cane sugar, fruit sugar (if present), water, and organic matter, or water and ash, as in the following TABLE V. -COMPOSITION OF RAW BEET-EOOT SUGARS OF GOOD QUALITY. Cane Sugar Per cent. Organic Matter. Per cent. Water. Per cent. 1. Clear light mixed product, . 85-0 1-6 20 2. Light mixed product, . 93-0 2 2-3 3. Half white, 1st product, 94-3 2-3 2-5 4. Light mixed product, . 93-5 2-4 3-3 5. „ „ „ . . . 94-4 31 "2 6. Half white, 1st product. 96-1 2-3 3-4 7. Clear white, 1st product, 92 3-7 2-8 8. Light mixed, 1st product. 93-5 3 2-8 9. Clear light mixed, 1st product, . 92 2-3 2-6 10. Clear light, 1st product. 93-1 3-3 3 11. Clear liiht, 1st product, 94-0 3-0 2-9 12. Clear yellow mixed product. 92 3 2-3 13. Yellow mixed product, 92 2-4 3-5 14. Clear yellow mixed product, 92 3 2 31 15. Clear yellow, 1st product, . 93-7 2-6 2-8 16. Yellowish mixed product, . 91-8 2-9 3-6 17. Yellow mixed product. 93-0 3-5 2-7 18. „ „ „ . . 93-4 2-9 2-3 19. „ „ „ . . 93-0 2-3 2-9 20. Yellowish 93 3-7 31 21. Y'ellow mixed product. 93 3-3 2-7 22. Light mixed product, . 91-6 3 3 23. Y^ellow „ 90-7 4-6 4-3 24. „ „ „ . . 93 -S 21 2-8 25. „ „ „ . . 92-5 2-4 3-5 26. Y^ellowish mixed product, . 93-0 4-3 3-7 27. Yellowish brown, 2nd product, 91-3 51 2-5 28. Yellow, 1st product, . 92-4 3-5 2-6 29. Yellow mixed, 1st product, . 89-5 4 3-7 30. Light yellow, 1st product, . 92-0 2-8 3-4 31. Yellow mixed product. 89-6 4-3 3-8 32. Brownish, 2nd product, 89 6-3 3 33. Second product, . 87-0 6-2 41 134 foods: their composition and analysis. [§69. TABLE VI, -SOME ANALYSES BY MR. HALSE OF CONCRETES. Cane Sugar. Per cent. Uncrystallised Su^ai-. Per cent. Water. Per cent. Ash. P er cent. L S7-20 4-00 4-50 1-38 2. 89 -GO 1-90 5-50 •85 3. 90 '20 1-95 4 03 •99 4. 91-70 3-30 2-15 •88 5. 87-00 5-00 4-72 1-24 6. 95-10 1-40 1-56 •86 7. 94-30 1-70 2-21 116 8. 92-50 1-92 2-70 1^18 The difference between the totals and 100 would be returned as " unestimated matters and loss." The methods of estimating the different kinds of sugar are fully considered in the next section, and it only remains to detail the best methods of taking the ash of a sugar. There are two methods of taking the ash of sugai*. The one is simply to burn the ash in the ordinary way in a platinum dish heated to redness in a current of air. In the case of all substances like sugar or starch, this method is very tedious, and without doubt there is some loss by volatilisation, Landolt* determined the amount of this volatilisation by a series of careful experiments, and gives the following Table (VII.), which may be used as a guide to the correction of the final weight of the ash. A method recommended and practised by Scheibler was to moisten the ash with sulphuric acid, whereby the combustion is much hastened, and the bases, being obtained as sulphates, approximate more nearly in weight to that of the organic salts naturally in the sugar, which in the other method are obtained as carbonates. It has also been proposed to precipitate the sugar with acetate of lead, and thus obtain the lead salts of the organic acids. The lead compounds are decomposed in the usual way, and the acids set free titrated by potash. The potash * Landolt, H., Journ. fur Prahische Chem., ciii. Also, "Sugar Cane," 1873, 70.] SUGAR. 135 combination appi'oximates somewhat more closely to tlie actual salts of the sugar. TABLE VII. Weight of Eesidue. Loss by Heating after Half an Hour One Hour. One and a Half Hours. Two Hours. •01 grm. •002 grm. •004 grm. •006 grm. •008 grm. •02 „ •002 „ •004 „ •007 „ •008 „ • •03 „ •002 „ •005 „ •OOS „ •010 „ •04 „ •003 „ •006 „ •009 „ •012 „ •05 „ •004 „ •007 „ •Oil „ •015 „ •06 „ •004 „ •oos „ •013 „ •017 „ •07 „ •005 „ •010 „ •015 „ •019 „ •08 ,, •005 „ •010 „ ■016 „ •020 „ But the best of those methods which attempt to reconstruct from the ash the original salts, is probably that of Laugier, already described at p. 123. Laugier extracts the organic acids by ether, and then adds thenr to the ash, and evaporates them down with it. As to raw beet sugar ash, the experiments of Landolt appear to show that simply multiplying the potassic carbonate found by 2, gives the amount of organic salt from which it was derived. * His experiment was as follows : — Two pounds of syrup were fully pi*ecipitated by lead acetate, then decomposed by SH^ and exactly neutralised by potash. The solution was next partly evaporated, passed through animal charcoal, and dried. It gave the reactions of chlorine, and of oxalic, malic, and tartaric acids, with a trace of sulphuric acid. Three separate portions were now carbonised, and the proportion for every one part of organic salt of carbonate of potash was — in experiment 1, 2'04; in experiment 2, 2-05; in experiment 3, 1^9S : the mean of the three being 2^08. § 70. Glucose, Dextrose, Dextro-glucose, Grcqje Sugar, CgHjoOgHoO — The rotatory power of glucose is 53°. It is soluble in lUO parts of cold water, and very soluble in boiling water ; it is * Or if Scheibler's process fee followed, the sulphates of the alkalies may be multiplied by 1-54. 136 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 7L soluble in glycerin, in about two parts of rectified spirit, and two of amylic alcohol ; but it is insoluble in ether and in chloro- form. Dextrose is widely spread in the vegetable kingdom, but is never found unaccompanied b}'- levulose. Dextrose is artificially obtained by heating carbo-hydrates, such as starch or cane sugar, with acids ; in such cases, it is accompanied by dextrin, from which it is difficult to purify it. According to Hoppe-Seyler, indeed, it cannot be obtained pure, save from diabetic urine, and the specific rotation iisually given is erroneous. He has separated pure grape sugar, dextrin free, from diabetic urine, and gives its polarisation as 53°"5. This, however*, agrees very nearly with that given by Tollens, who ascribes to anhydrous dextro-glucose a specific rotation of 53°-l, and to the dextrose with its water of crystallisation a specific polarisation of 48°-27. The best way to obtain dextrose from cane sugar in a pure state is, according to Soxhlet, the following : — 3 litres of 90 pei\ cent, alcohol and 120 cc. of concentrated hydrochloric acid are made to act at 4.5° for two hours on 1 kilo, of cane sugar. After ten days, crystals of dextrose form, wlien the liquid may be concentrated by distillation, and the crystals which have formed removed. In a few days, the whole of the dextrose will have been deposited as a white powder. The crystals are washed with 90 per. cent, alcohol and with absolute alcohol, and crystal- lised out of the purest methyl-alcohol. Crystallised grape sugar is in the form of little masses of six-sided tables, which melt at 86°, and lose at 100° their water of crystallisation. § 71. Levulose (or Levuglucose) is isomeric with dextrose, but distinguished from it by its action on a ray of polarised light — turning to the left, instead of to the right: - 106° at 14"", - 53° at 90°. It is obtained in company with dextrose when sugar is "inverted" by the action of a dilute acid. To isolate levulose the acid must be got rid of; for example, if hydro- chloric acid has been used, it is precipitated by silver solution; if sulphuric, by baryta water, &c. The solution of invert sugar must be about 10 per. cent, strength. To every 100 cc. 6 grms. of freshly burnt lime must be added, and the whole shaken. By artificially cooling the solution with ice, a crystalline magma is obtained, and by filtration the more soluble dextrose lime- compound can be obtained from the less soluble levulose lime- compound. The sugar thus obtained can be freed from lime by carbon dioxide. Levulose is uncrystallisable, but it , has not been found possible to separate it entirely from the crystalline glucose, by crystallising the latter out of it. It presents when pure simply the characters of a coloui-less syrup. § 72.] ESTIMATION OF SUGAR. 137 ESTIMATION OF SUGAR. § 72. Sugar is estimated by chemical processes, by the specific gravity of the solution, by the estimation of the COo evolved in alcoholic fei-mentation, and by certain physical processes. It is only possible to estimate percentages of sugar (especially cane sugar) from the specific gravity of the solution when the fairly pure sugar is dissolved in pure water, so that this method is of but limited utility, and seldom employed by the analyst. (1.) Chemical Processes depending upon the Precipitation of the Suboxide of Copper from a Copper Solution hy Grape Sugar. The most general of the numerous processes under this head is that of Fehling, which requires a solution of cupric sulphate and Rochelle salt, alkalised by soda. 34-64 grms. of pure crystal- lised cupric sulphate, previously powdered and pressed between blotting-paper, are dissolved in 200 cc. of distilled water ; 174 grms. of Rochelle salt are dissolved in 400 cc. of a solution of pure caustic soda, specific gravity 1-14; the two solutions are now mixed and made up to 1 litre. The liquid should be preserved in bottles protected from the light, and absorption of carbon dioxide from the air should be provided against. On account of the slight instability of this solution, A. Soldaini has proposed the following : — 416 grms. of potassium bicar- bonate, 15 grms. of dry basic cupric carbonate, and 1400 grms. of distilled water ai*e heated together, the liquid being continually replaced ; when the evolution of COj has nearly ceased, the liquid is made up to its original volume with water, filtered, and concentrated down to 800 cc. Such a solution is not reduced by light or the carbon dioxide of the air : it is unaltered by pro- longed boiling, and may even be evaporated to dryness without decomposition. It is reduced by formic acid, levulose, glucose, and lactose, and can be used for quantitative pui-poses in the same way as " Fehling." Cane sugar cannot be estimated directly by " Fehling," since it does not redvice copper solution ; by boiling with dilute acid it is, however, changed to inverted sugar, Avhich reduces copper or mercury in nearly the same proportion as glucose (see p. 140). Starch and starchy substances may be also changed into sugar by boiling for several hours with a dilute acid. The following is the method used by the author : — 138 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 72. The starcliy substance is purified by boiling with a mixture of alcohol and ether to remove fat ; it is next digested with cold water to remove sugar and soluble albuminoids; the purified starch is dried, and a grm. is heated with 50 cc. of water in the water-bath for an hour; to the hot solution a grm. of oxalic acid in 25 cc. of water is added, and the mixture heated for two hours ; at the end of which time the liquid is cooled, made tip with water to 100 cc, filtered, and an aliquot part examined in a Laurent's or other saccharimeter, and also the reducing power ascertained by means of the cyanide copper process (p. 143). The degrees of rotation which the sugar found should produce are subtracted from the total rotation, the result being the rota- tion due to dextrin ; the sugar found is then multiplied by 0*9 for 100 parts of sugar = 90 starch, the amount of dextrin added — and the result = starch. (2.) Volumetric Processes hy the aid of Solutions of certain Salts of Mercury. Knapp's mercuric cyanide solution is made by dissolving 10 grms. of mercuric cyanide in about 600 cc. of water, then adding 100 cc. of caustic soda solution of sj^ecific gravity 1-145, and dilut- ing to 1 litre. 40 cc. of the mercuiy solution are placed in a flask, heated to boiling, and the solution containing sugar run in gradually from a burette, four parts of me^'curic cyanide being reduced to metallic mercury for every one part of anhydrous grape sugar (or, 3-174 parts of metallic mercury = 1 anhydrous grape sugar). The ending of the process is discovered by moistening filter-paper with the clear solution, and holding quite close to it a rod dipped in ammonium sulphide solution ; a decided brown coloration takes place if the mercury salt is in excess ; but if the colour is very faint, the operation is finished, for it appears to be impossible to decompose the whole salt, a trace always remaining, and for this reason the solution should be standardised with sugar. A. Sacchse uses the following solution for the estimation of sugar: — 18 grms. of pure dry mercuric iodide, and 25 grms. of potassic iodide are dissolved in water, a solution of 80 grms. of caustic pota.sh is r.dded, and the whole made up to 1 litre. 40 cc. of this solution [ = 0-72 grm. Hglg] are boiled in a basin, and the solution of grape sugar is run in, until the whole of the mercury is precipitated. The final point is determined by spotting a drop of the supernatant liquid on a white slab, and then bringing it into contact with a drop of a strongly alkaline solution of stannous chloride. The production of a brown colour shows the presence of unprecipitated mercury. § 73.] ESTIMATION OF SUGAR. 139 § 73. Some researches of F. Soxblet* bear upon the behaviour ©f the diflerent kinds of sugar with the solutions just described. These experiments are of the more importance since they •were made with the most scrupulous care, and with the purest materials, while the testing was done under varied con- ditions. Invert Sugar. — Pure invert sugar he prepared by dissolving cane sugar which had been purified by thrice crystallising out of water, and dried at 50° in a vacuum over calcic chloride. 9-5 grms. were dissolved in 700 cc. of boiling water, to which had been added 100 cc. of hydrochloric acid, containing -72 of HCl, and heated in the water-bath for half an hour. The liquid was finally neutralised Avitli soda, and diluted to one or two litres as might be required. He found that invert sugar reduced Fehling's solution almost at once, and that in all cases two minutes were sufficient, no advantage being gained from a longer boiling; nor did it appear to matter whether the sugar was added to the liquid cold, and then boiled up, or the sugar added to the boiling liquid. Soxhlet drew from this experiment the following conclusions :— 1. The pi-oportion in which invert sugar reduces copper oxide is essentially influenced by the concentration of the liquid. Dilution of the copper and sugar solution lowers, excess of copper raises, the reducing power. "5 grm. invert sugar in 1 per cent, solution corresponds to 101-2 cc. of undiluted Fehling's solution ; but, if Fehling is diluted with four times its volume of water, then the same amount of invert sugar is equivalent to 97 "0 cc. of Fehling: in the one case the proportion in equivalents being as 1 : 10-12, in the other as 1 : 9-70. 2. In the titration of an invert sugar solution, the first cc. of the sugar solution flowing into the copper reduces more copper than the next, and the last cc. has the smallest reducing power, because the first has the gi^eatest excess of copper solu- tion to act upon, and the last the smallest. It hence follows that the reduction proportion is not constant throughout the operation, but is continually falling, and that the values are purely empirical, and only correct by operating always with the same concentration of copper and sugar solution. 3. The accepted view that 1 equivalent of invert sugar reduces 10 of CuO is wrong. -5 of invert sugar does not reduce 100 cc. of Fehling's solution diluted with 4 of water; 97 or 100 cc. of Fehling are not equivalent to -500 grm., but to -515. Milk-Sugar. — Similarly with regard to milk-sugar, he gives * Joiirnal f. Prakt. Chemie, KF. xxi., 227, 317 ; Zeitschrift. f. Analijt. Chemie, xx. 425. 140 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 73. the reduction proportion in equivalents as 1 is to 7 '40, or 0'5 grm. of milk-sugar in 1 per cent, solution is equal to 74 cc. of Fehling's solution. Dilution of the solution, or concentration, has a similar action to that of invert sugar, but smaller in degree. Milk-sugar solutions require six minutes' boiling. Galactose. — Galactose reduces Fehling as quickly as invert or grape sugar. "5 of galactose is equivalent to 98 cc. of vindiluted Fehling. If the latter be diluted with four times its volume of ■water, then it corresponds to 94 cc. The reduction ratio in equivalents is as 1 : 9-8, and as 1 : 94 in the respective cases mentioned. Maltose. — The behaviour of maltose, according to Soxhlet, is as follows : — Maltose has the smallest reducing power of all the sugars. It reduces the copper solutions more slowly than grape, invert, and galactose, but more rapidly than milk sugar. -5 grm. of maltose in 1 per cent, solution equals 64-2 undiluted Fehling ; and the same quantity corresponds to 67-5 Fehling, if the Fehling is diluted with four times its volume of water. It is remarkable that dilution of the solutions increases the reducing power ; while, with regard to undiluted Fehling, excess of copper appears to have no influence. This fact affords an easy way of estimat- ing maltose by weight ; for there is, under these circumstances, and operating with 1 per cent, maltose, only one ratio, viz. — 100 of anhydrous maltose equalling 11.3 copper. It is, of course, necessary to make sure that the copper solution is in excess. The fluids are mixed cold, boiled for four minutes ; the sub- oxide collected on an asbestos filtei-, reduced in a current of hydrogen, and weighed as metallic copper ; or it is also open to the analyst to redissolve the copper from the filter by an acid, and precipitate on a platinum dish by electrolysis. Soxhlet has also investigated the behaviour of the solutions described with various sugars, and gives the number of cc. of the quicksilver solution reduced by 1 grm. of the difterent sugars when dissolved so that the solutions are of 1 per cent, strength. Knapp. Sacchse. Grape sugar, 497 '5 cc. 302 5 cc. Invert sugar, 502-5 376-0 Levulose 50S-5 449-5 Milk-sugar, 322-5 214-5 Galactose, 413-0 226-0 Changed milk-sugar 448-0 258-0 Maltose, 317-5 197-0 Or, if the reducing power of grape sugar be taken as 100, then the other sugars may be thus compared — ^ 73.] ESTIMATIOX OP SUGAR. 141 Fehling undiluted. Knapp. Sacchse. Grape sugar, 100-0 cc. 100-0 CC. 100-0 cc. Invert sugar, 96-2 99-0 124-5 Levulose, . 92-4 102-2 148-6 Milk-sugar, 70-3 64-9 70-9 Galactose, . 93-2 83-0 74-8 Changed milk-sugar. 96-2 90-0 85-5 Maltose, 61-0 63-8 65-0 As to the recognition of the kind of sugar, Sacchse's solution differs in its behavioixr with the various kinds more than Knapp's solution, and is, therefore, the best adapted for this purpose. These sohitions possess no advantage over Fehling's in the •estimation of sugar, if one kind only is present ; but where there are two kinds of sugar in one sample, or, again, where the identity of the sugar is doubtful, then the mercviry methods are of very great use and importance. Sacchse recommended the ■use of his method, combined with that of Fehling, to determine the relative proportions of grape and invert sugar in a mixtui'e. He considered that gi-ape and invert were reduced by Fehling in exactly equal proportions, and by his mercury solution in unequal proportions ; but the researches of Soxhlet just detailed show that Fehling does not act in the way assumed by Sacchse, so that the calculation for the amount of sugar in a liquid by the combined method is somewhat different, and must be done according to the following equations. The general formula is — ax + by = F, ex + dij = S, a is the number of cc. Fehling's solution, reduced by 1 grm. of grape sugar. b is the number of cc. Fehling's solution, reduced by 1 grm. of invert sugar. c, the number of cc. Sacchse's solution, reduced by 1 grm. of grape sugar. d, the nu.mber of cc. of Sacchse's solution, reduced by 1 grm. of invert sugar. F, the number of cc. of Fehling's solution, used by 1 volume of sugar solution. S, the number of cc. of Sacchse's solution, used by 1 volume of sugar solution. X, the amount of the grape sugar in grammes contained in a volume of the sugar solution. 142 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 73. y, the amount of the invert sugar in grammes contained in a volume of the sugar sohition. Since, with the titration of the mixed sugar solutions, equal amounts of the mercury and copper solutions are taken, which use up unequal amounts of the sugar solution, it is most con- venient to calculate the number of cc. of Sacchse and Pehling equivalent to 100 cc. of the sugar solution. Putting Soxhlet's values in place of the symbols, the equa- tion for grape and invert sugar becomes as follows : — 210 -40; + 202-4?/ = F 302 -Sx + 376-0^ = S Or if the mixture is grape sugar and galactose, the equation is — 210-4a; 4- 196% = F 302 -52; + 22G% ^ S Therefore, since the F and S are found experimentally, then there are only two unknown quantities, which are easily cal- culated by the ordinary rules. It is scarcely necessary to ob- serve that the same calculation applies to Knapp's solution, if the proper valvies are substituted in the formula.* Soxhlet's method of using Fehling's solution is as follows : — 50 cc. of Fehling's solution are heated to boiling, and the sugar solution run in, in the usual way, until the blue colour disappears. This gives the approximate strength of the solution, and it must be now diluted so that there will be about 1 per cent, of sugar in the solution. A second 50 cc. are now taken and heated with the exact number of cc. of the solution, which (supposing it to be accurately 1 per cent.) would thi-ow down all the copper. This heating is to occupy two minutes for invert sugar, grape sugar, and lactose ; four minutes for maltose ; and six minutes for milk-sugar. The whole fluid is now poured on to a large » 100 cc. of Knapp's solution are reduced by 100 cc. of Sacchse's solution, are reduced by Anhydrous. Milligrammes. Milligrammes. Grape sugar, 2010 330-5 Invert sugar, 199-0 266 Levulose, . 197-0 222-5 Milk, 310 466 Lactose, 242-0 442 Maltose, . 315 5060 r§ 73. ESTIMATION OF SUGAR. 145 filter ; if the filtrate is greenisli, copper is of course present; but if it is yellow still there may be copper dissolved, and a little must be tested in a test-tube with acetic acid and ferrocyanide of potash solution. A dark red colour shows a large amount, a pale red a small amount ; but if there is no colour at all the copper is precipitated, if copper was in the solution. In the next experiment a slightly larger amount of sugar is used, but if free from copper then in the next assay 1 cc. of sugar solution less is taken. These titrations (which are very rapidly executed) ai-e continued imtil in two experiments the addition or subtrac- tion of 1 cc. gives, on tlie one hand, a copper-free, and, on the other, a trace of copper-containing liquid. In dark liquids the ferrocyanide and other tests of the kind are unsuitable ; but iu such a case a few drops of the filtrate are put in a test-tube, boiled with a little sugar solution for a minute, and then put on one side to deposit for two or three minutes. The fluid is now decanted, and a little })iece of white filter-paper, which has been previously wound round a glass rod, wiped around the bottom, when any oxide of copper which has been deposited adheres to the paper in this way, and is at once discovered. The Cyanide Copper Process."^ — This process is convenient and accurate, and is performed as follows : — To 20 cc. of Fehling (or, as it is better to keep the solutions separate, to 10 cc. of copper sulphate solution of Fehling's strength, to which 10 cc. of the alkaline tartrate are added), add 40 cc. of water in a porcelain dish, and heat to boiling ; while boiling drop in from a burette a 5 per cent, solution of polassic cyanide until there is only the faintest blue colour observable — now add a second 20 cc. of Fehling, and heat to boiling, and while boiling drop in the solution of sugar from a burette until complete decolorisa- tion. The liquid should be standardised by solutions of 0-5 per cent., 1-0 per cent, and 2 per cent, of the various reducing- sugars. Mr. W. A. Rogers has n:ade a number of determinations of lactose in milk by the copper cyanide ])rocess in my laboratory, and has observed that when tlie end of the reaction is nearly reached, it is necessary to add the sugar solution very carefully, since the addition of any excess of the latter, more than neces- sai'y to just decolorise the solution, produces a green colour which may rather easily be mistaken for the faint blue which precedes the end of the reaction. In making a number of deter- minations, it is also advisable to allow the same length of time (say a minute) to elapse between the addition of the portions of the sugar solution ; otherwise strictly comparative results may not be obtained. * Gerrard, Pharm, Journal, 3d series, xxv. 912. 144 FOODS : THEIR COMPOSITION AND ANALYSIS. (3.) Dr. Pavifs Process. An excellent method of determining sugar lias been invented by Dr. Pavy.* The principle of the process depends on the decolorisation of an ammoniacal copper solution by glucose in the absence of air. The copper solution is made by dissolving 204 grms. of potassic sodic tartrate and 20-4 grms. of caustic potash in 200 cc. of water, in another 200 cc, 4-158 grms. of cupric sulphate are dissolved by the aid of heat, and the two solutions are mixed together ; when cold, 300 cc. of strong ammonia (sp. gr. '880) are added, and the whole made up to 1 litre with water. 10 cc. of this liquid diluted to 20 cc. with water equals 5 mgrms. of glucose. To make the estimation, 10 cc. of the ammonia copper solution are placed in a flask with a rather wide mouth, an equal bulk of water is added, and a good caoutchouc stopper having two per- forations is fitted to the neck of the flask, tlirough one of the holes is adapted air-tight the conical end of a burette, having either a glass stopcock or a clip ; while the other hole carries a bent tube for the exit of the vapour. The sugar solution is placed in the burette. Heat is now applied to the flask, and when the liquid is boiling violently the sugar solution is run in slowly, after each addition boiling up ; the blue colour fades gradually, and the end reaction is the complete absence of blue colour. The author uses a modiflcation of the process invented by Mr. Stillingfleet Johnson. The caoutchouc stopper is perforated by three holes instead of two ; the third is for a tube which, dips beneath the surface of the copper solution, and is closed or opened at pleasure by means of a short bit of india-rubber tubing adapted to the air-end of the tube and furnished with a clij) ; the second tube, for the outrushing ammonia vapour and steam, is also furnished with an india-rubber tube, the end of \vhich dips under the surface of a considerable bulk of acidulated water, and is furnished with a Bunsen valve to prevent any back-rush. The use of the latter tube is, of course, to condense the ammonia-vapour, so that the operator is not inconvenienced. The use of the extra tube is to more accurately hit the end reaction — to do this, directly the decolorisation of the copper solution is complete, the flame is removed and the clip opened ; as the flask cools air passes in a stream of bubbles through the liquid, if the point has been exactly reached, the blue colour * Lancet, 1884. § 73.] ESTIMATION OF SUGAR. 145 reappears after a very few seconds, but if, on the contrary, too much sugar solution should be run in, a longer time elapses.* By standardising the copper solution by a pure glucose, and working so that after complete decolorisation a certain number of seconds elapse, before the blue colour reappears, and taking with unknown solutions the same number of seconds, a high degree of accuracy is attainable. It is, of course, obvious that the sugar solution to be tested must be very dilute — viz., from -4 to -8 per 1,000. A preliminary experiment must first be made, and then the solu- tion so diluted that from 6 to 10 cc. are required to decolorise 10 of ammoniacal copper. The following table, taken from Dr. Pavy's original paper, may be useful : — TABLE VIII.— Showing the Amount of SuCxAR expressed in Parts by Weight per 1000 by Volume corresponding with cc. in IOtiis REQUIRED to DeCOLORISE 10 CC. OE THE AMMONIATED COPPER TeST. Cc. to Decolorise. Parts per 1000 of Sugar. Cc. to Decolorise Parts per 1000 of Sugar. Co. to Decolorise. Parts per 1000 of Sugar. 6 •833 8-0 •625 9^1 •549 6-1 •819 8-1 •617 9-2 •543 6-2 •806 8-2 •609 9-3 •537 6-3 •793 8-3 •602 9-4 •531 G-4 •781 8^4 •595 95 •5^26 6-5 •769 8-5 •588 9-6 •520 6-G •757 8-6 •581 9-7 •515 6-7 •746 8^7 •574 98 •510 6-8 •735 8^8 •568 99 •505 6-9 •724 8-9 •561 10 •500 7-0 •714 9 •555 10 1 •495 (4.) Physical Processes for the Determination of Sugar. The saccharimeters in use are numerous ; for the food analyst, the most useful are the larger instruments, which admit of the use of tubes up to 500 mm., for in this way, and in this way only, can solutions containing 0-5 per cent, of sugar be physically estimated. Mitscherlich's polariscojye (.see fig. 20) consists of a stationary Nicol's prism in a, a plano-con- vex lens in h, and a rotating Nicol's prism c. The first prism * It has been proposed to exclude the air by a layer of parafiBn of high boiling point. 146 FOODS : THEIR COMPOSITION AND ANALYSIS. [§73. polarises the light, and the use of the second is to indicate the plane of the polarised ray coming from the first. The second prism is therefore set in a graduated circle, dd, and is provided with an index, /, and there is a handle, e, which turns both prism and index. If the index be either at 0° or 180°, and an observer look through the tubes towards the source of light, the flame is seen divided by a vertical line into two equal parts ; if now the tube, sup- plied with the instrument, be filled with the liquid to be examined, and interposed between the lens and the second prism, should it con- tain sugar or other polarising substance, the black stripe is no longer in the middle of the field, and the handle moving the index and prism must be turned until the black stripe is seen; or, if the stripe is broad and undefined, the prism is turned until the exact point is reached in v/hich blue changes into red — the index at this point marking the amount of the polarisation by the scale and the direction ; for if the index has to be turned to the right, the polarisation is + , or right-handed; if to the left - , or left-handed. In order to make this quantitative, and to estimate the specific rotation of a sugar (i.e., the number of degrees of I'otation observed when 1 grm. of the sugar is dissolved in 1 cc. of fluid and observed by yellow light through a tube 1 decimetre long), it is necessary to dissolve a known weight of the pure sugar in water ; then if the length of the tube be known, and the temperature of the solution and the rotation be observed, all the necessary data are obtained. For example, let the rotation = a, the length of the tube in decimetres = 1, the weight of substance in 1 cc. of fluid = 2^} then the specific rotation for yellow light — ; Fig. 20. p\ or (a)j pi the sign (a) ; being in use, signifying yellow light. Or, to take an actual example : 14-3 grms. of a substance dissolved in 100 cc. § 73.] ESTIMATION OF SUGAR. 147 (■143 grm. in each cc), and a 2 decimetre long tube filled with this liquid, the rotation on the scale being IG'' to the right, then ^I^ = ^^•^^' and 55-94 is the specific rotation. The best source of light for accurate researches is a Bunsen burner, in the middle of which there is a little pellet of sodium held on a wire. This source of light in fcrmulaj is usually indicated by (a)^. Provided there be only one su.gar in the fiuid under investiga- tion, the specific rotation of which is known, the weight of the sugar in 1 cc. of the fluid is estimated by the following formula : P = , where a equals the observed,* and (a) the specific rotation. Soleil's Saccharimeter {see Plate) consists of three essential parts, two of which are fixed (fig. I), AB and CD, the other movable, which is inserted between B and C, and which is some- times the tube BC, 20 centimetres long (fig. 2), and sometimes the tube B'C (fig. 3), 22 centimetres long, furnished with a ■ thermometer, T. These tubes are destined to contain the saccharine solutions, the value of which is to be determined. The movable parts are — (1.) The small movable tube D'D (fig. 1), carrymg the eyepiece, which focuses by drawing in and out. (2.) The little button V (fig. 4), serves to adjust the zero of the scale with the zero of the indicator. (3.) The large milled screwhead on the vertical axis H (fig. 1), by which is rendered uniform the tint observed. (4.) The milled ring B (figs. 1 and 2), by the aid of which they give to this same tint the colour which lends itself best to a precise valuation. (5.) Lastly, the divided scale RIl (fig. 4), on which is read the number giving the richness of the sugar under examination. The details of operating are as follows : — The lamp is adjusted so that its light traverses the axis. A tube similar to that which contains the saccharine solution is filled with pure water, and is adjusted in the place provided for it between the ocular and objective portion. Then a]Dplying the eye at D (fig. 1), the tube DC is either pushed out or in, until the field is seen divided into two equal halves, coloured with one and the same tint, or two diff'erent tints separated from each other by a black line, which should be very sharply defined. If, as generally * That is, the observed rotation in a 1 decimetre tube, but if, as is usual, the tube be 2 decimetres long, the observed rotation must be divided by 2— e.g., the rotation in a 2 decimetre tube being as above 16°, the weight of the sugar in each cc. is '143 grm. for 148 FOODS : THEIR COMPOSITION AND ANALYSIS, [§ 73. happens, the two half-discs have not the same tint or shade, the large horizontal button H is turned either way until the desired result is obtained. It is not only necessary that the two half-discs should have the same tint or colour, but in order to be extremely exact, that tint should be the one most sensible to the eye of the observer; and as all eyes are not equally sensible to the same tint, the proper colour must be found by experiment. The zero line on the scale must coincide exactly with the black line of the indicator I (fig. 4). If the coincidence is not perfect, it may be established by turning either way the little button V until this is accomplished. The instrument once adjusted, the examination of the sugar may be commenced. The tribe BC, filled with the saccharine solution, is substituted for that filled with water, or if an inverted sugar is taken, then B'C is filled. On now looking through the instrument, it is seen that uniformity of tint no longer exists, and that the two half-discs are coloured by different shades. The unifoi-mity is re-established by turning the large horizontal button H until the two half-discs are again uniform. As the saccharine solution is mostly coloured, the uniform tint re-established is not in general the sensible tint, to which, however, it is necessary to return, and which the colour of the solution has caused to disappear. The milled head B is then turned to cause the sensible tint to reappear ; this tint returned, the equality of shade of the two half-discs, if not quite perfect, must be made so by turning again H. It now only remains to read the degree on the scale RR', to which the index answers ; the number corresponding to this degree gives immediately in lOOths the titre, or the richness of the sokition. The preparation of the saccharine solutions is as follows : — (1.) Normal solutions of pure sugar. — 16-26 grms. of pure sugar are dissolved in water, the volume made up to 100 cc, and observed in a tube 20 cms. in length; marks 100 degrees on the saccharimeter. (2.) The raw sugar of commerce. — 16-26 grms. of the sugar are powdered and dissolved in water, and the whole made up to 100 cc. ; if the solution is too dark, it may be clarified by sugar of lead. The tube BC is filled with the solution thus prepared and adjusted. (3.) The next operation is to invert the sugar. 5 cc. of fum- ing HCl are added to 50 cc. of the sugar solution, and heated in the water-bath up to 68° ; when that temperature is reached, the solution is put in the tube B'C, its rotating power (which § 73.] ESTIMATION OF SUGAR. lid is now inverse) observed, and at the same time the temperature at the moment of the observation.* We have now all the data necessary, and the amount of sugar may be readily found by tables, such as those of M. Clerget, or by the formula as below. Supposing the nvimber given b}'- the first observation is 75, by the second (inverted) 21, at a temperature of 12°, the sum of the two numbers (75 + 21) makes 96. Now, in referring to M. Clerget's table, under 12°, or in the third column corre- sponding to the temperature of 12°, the nearest number to 96 is in this instance 95*6 ; the hoi'izontal line in which 95'6 is placed is followed, and there is found, first, in the column A, the figures 70 per cent, of pure crystalline sugar ; secondly, in the column B, the figvxres 1 14*45, placed by the side of 70, which indicates that the sacchaiine solution examined contains per litre 114-45 gi'ms. of pure sugai\ If, however, as sometimes happens, the sokition contains a polarising substance not modified by acids, in such a case the diffei-ence of the two numbers, and not the Slim, is to be taken and dealt with as before. It is scarcely necessary to remark, that if the substance is known to contain only crystallisable sugar, and the tube BC be filled, one observa- tion alone sufiices. If tables are not at hand, the following formitla can be used : — Let T be the temperature, S the sum or difterence of the two determinations, P the rotatory power, R the quantity of sugar contained in 1 litre of the solution : — _ 200S -, P X 16 -.350 _ ^ , ^.,. ^ = 288^T ^ = lO ^ "" ^■^''^ §'''"'• Professor Jelletfs instrument is a little more elaborate than Soleil's, and of gi-eat accuracy. The eyepiece or analyser of the apparatus consists of a suitably mounted prism, made from a rhombic prism of Iceland spar. The rhombic prism is cut by two planes perpendicular to the longitudinal edges, so as to form a right prism. The prism is next divided by a plane pai-allel to the edge just produced, and making a small angle with the longer diagonal of the base. One of the two parts into which the prism is thus divided is then reversed, so as to place the base upper- most, and the two parts are connected together. * The figures usually accepted for the rotatory powers of the sugars at 17*5° are as follows : — Lactose, . . + 52*5 Maltose, . . +138-3 Saccharine (Peligot), + 198-4 Cane sugar, , . + 66-5 Invert sugar, . - 22-4 Glucose, . + 53-0 Levulose, . - 1000 150 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 74. Another distinctive feature of the instrument is, that the mechanical rotation of the analyser for the finding of any par- ticukr plane is dispensed with, this function being transferred to a fluid which has the power of turning the plane of polarisa- tion opposite to that of the solution to be examined. The analysing tube slips into, and moves up and down in, the compen- sating fluid, so that diff'erent thicknesses of the latter fluid can be readily interposed and measured by a scale fixed to the instrument. The Saccharimetre ci Penomhres, of which the principle was enunciated by M. Jellet, as constructed by M. Duboscq, has some very great advantages. It requires the employment of a simple light, and the field does not present to the eye for com- parison two diflferent colours, but two intensities, sensibly diverse, of one and the same colour, so that the least variation can be appreciated. The simple light is best obtained by the insertion of a bead of some salt of soda on a platinum wire in the flame of a Buusen burner. Landolt's with Li])pich's pohiriser and Lau- I'eiit's half-shadow are all excellent instruments, and suitable for the purposes of the analyst. COXFEGTIOXERY,— SWEETMEATS. § 74. It would take many pages to describe the composition of the various kinds of sweetmeats in commerce : the basis of all is either cane or grape sugar, or honey, flavoured with appropriate essences, and coloured with various colouring matters. A great many common sweetmeats have a most definite composition, and it is evident that a deviation from the ordinary process of manu- facture must, if it should take the form of substituting inferior ai'ticles for, or the addition of matters giving weight to, that which is ordinarily sold, would be an adulteration. As an example "peppermint lozenges," or "peppermint drops," are composed of albumen, cane sugai", and oil of peppermint. None of these ingredients have any amount of mineral matter, and peppermint lozenges, when burnt, do not leave as much as -2 per cent, of ash. Since they are sold by weight it is easy to adulterate them by mineral substances ; but such an addition would be most decidedly fraudulent, and the analyst may justly certify accordingly. A large proportion of the common sweets contain nothing else besides sugar, for the manufacturer, by careful heating, is able to impart a quite surprising scale of colours, from the purest white to fawn colour, straw colour, reddish-brown, brown, to almost a jet black, by this agent alone. § 75.1 CONFECTIONERY, — SWEETMEATS. 151 Sugak-Candy is simply crystals of sugar obtained in a particu- lar way, and is of all colours — from the white candy, largely used for the manufacture of artificial champagne, to all shades of yellow and red. As usually manufactured, the purified sugar solution is concenti'ated to a specific gravity of 1'420 to I'-iSO, and then run into copper cones, through which are passed a number of threads ; these cones are heated with warm air, and the crystallisation occupies as much as from eight to fourteen days. The composition .of white candy, made from pure loaf- sugar, is as follows : — Percent. Crystallisable sugar, . . . . 80-00 Uncrystalhsable sugar, . . . Traces. Ash, 0-0 Water, 20 00 The coloured candies may contain some mineral matter, and a good deal of uncrystallisable sugar ; copper may, as an impurity, be present. Co7)iposition of Sweetmeats Generally. Toffy. — Toffy is made by melting sugar with butter, and etlier will extract a large amount of fat. The ice-coating of cahes is composed of white sugar and albumen. A 'great many sweets are acidulated with citric acid, and a few have cavities within them, supposed to contain alcohol, but really a little syrup. Gum, tragacanth, citric acid, fruit sugar, gelatin, albumen, fatty and flavouring matters, with the follow- ing colouring-matters, make up the usual harmless ingredients of the confectioner's shop : — Red. — Cochineal, carmine, the juice of beet and of red berries, such as cherries, currants, &c. Yellow. — Saflron, safflower, turmeric, marigold, Persian berries. Blue. — Indigo, litmus, safi"ron blue. Green. — Spinach juice and mixtures of yellow colours with blue. Black. — Chinese ink. Besides these, there are the aniline colours, which, when pure, Lave not been proved to be injurious. Analysis of Sweetmeats. § 75. The analyst will naturally first turn his attention to the percentage of sugar, and estimate the total amount in the usual way; and, if necessary, investigate by optical and chemical means, whether there is more than one kind of sugar present. The essential oils may be dissolved out by petroleum ether, and identified by their odour ; but the colouring-matter will, for the most part, be the chief substance necessary to examine. If the 152 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 75. colouring is only on the external surface, it is better to detacli it by scraping or rasping, than to powder the whole substance up, for if the colour is carefully detached as pure as possible, tests may sometimes be directly applied without any further trouble. The colour by treatment with alcohol, with water, and with bleaching powder, is quickly referred either to the organic or to the indrganic division of chemical substances. With regard ta organic colours generally, the reader may consult the sections treating of " Colour," where full directions are given for their identification. If, however, the colour is apparently inorganic, then the following substances may be particularly tested for :-^ Among RED colours — iron ; „ YELLOWS — chromate of barium, and lead compoundSf arsenic and antimony ; „ GREEN — arsenic, copper ; „ BLUE — Prussimi blue ; „ WHITE — sulphate of barium, salts of zinc. A weighed portion of the scraped-off colouring-matter is burned to an ash, which is dissolved in hydrochloric acid, and tested with hydric sulphide, after adding just sufficient soda to so neutralise the acid as to leave only a slight excess. Under these circumstances, lead, copper, or zinc, if present, will be precipi- tated ; while, if it is strongly acid, zinc would remain almost entirely in solution. Ammonium hydrosulphide is next added to the solution, which has been boiled and filtered from any pre- cipitate ; this reagent will throw down iron, manganese, &c. To test for chromium, it is best to boil the colouring-matter with a sohition of carbonate of potassium, when potassic chromate will be formed, which gives, in neutral solutions, a purplish precipi- tate with nitrate of silver. Barium is easily detected by fusing the ash with carbonate of soda, dissolving the ash in dilute hydrochloric acid, and adding a little hydric sulphate ; a heavy characteristic precipitate of barium sulphate is thrown down. If barium is present, it may exist with evidences of chromium, in which case, in all probability, the colouring-matter was chromate of barium, or if the sweetmeat is not coloured by barium chromate, baryta sulphate may have been added simply to give weight. Arsenic and antimony are best discovered by boiling a little of the colouring-matter with copper-foil [Reinsch's test] ; and although this test will not detect quite such a minute quantity as Marsh's test, it is sensitive enough. Copper is also best detected by electi-olysis, the substance being placed in a platinum dish, acidified, and then a rod of zinc inserted ; or, the neater plan of connecting the dish itself with a battery may, where appliances are at hand, be preferred. 76.] HONEY. 153 HONEY. § 76. Commercial honey is the saccliarine matter collected and stored by one particular species of bee (Apis mellijica) ; but the production of honey is by no means limited to the bee, for there is a honey-ant* in Mexico, which stoi'es a nearly pure syrup of uncrystallised sugar. This is slightly acid in reaction, and reduces salts of silver like formic acid.t From determinations of the amount of saccharine matter in different flowers, it has been calculated that to make 1 kilogramme of honey, the bees must visit from 200,000 to 500,000 flowers. A wasp of tropical America is said to yield a honey in which are found crystals of cane sugar, but the evidence as to this latter point is not decisive.;]; A curious sample of honey has been analysed by A. Villiers.§ It was derived from Ethiopia, and is the produce of an insect resembling a large mosquito, which, like our wasp, makes its nest in cavities in the ground. It secretes no wax. The natives call the honey " tazma," and ascribe to it medicinal virtues, especially using it as a cure for sore throat. Its composition is as follows: — Per cent. Water, 25-5 Fermentable sugar (levulose with a sixth of glucose in excess), 32 Mannite, .......... 3'0 Dextrin, .' , 27-9 Ash, 2-5 Loss and unestimated, . . . . . , . . 9'1 The honey contained a non-nitrogenous bitter principle. The essential constituent of honey is a mixtui-e of dextrose and levulose ; it also contains mannite, wax, formic and other organic acids, pollen, not unfrequently alkaloidal and bitter principles from the plants, possibly derived from the pollen, small quantities of cane sugar, of mineral matter, and invariably minute quantities of alcohol. The properties of dextrose and levulose have been already described. The other saccharine constituent of honey — mannite, CgHj^Og — crystallises in four-sided prisms, is soluble in 80 parts of alcohol of specific gravity 0-898, and in 1-400 parts of absolute * The Myrmecocysttis Mexicanus. There are two kinds of workers— one the active form, the other sedentary— which produce the honey. The latter is the larger, and has a tumid abdomen ; it never quits the nest. The honey is discharged into proper receptacles, and from it the Mexicans make a pleasant di'ink. t H. Marsten, Pogg. Ann., c. 550. t G. M. WethereU, Chem. Gaz., 1S53, 72. § Compt. Rend., Ixxxviii. 292, 293. 154 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 7G. alcohol ; in boiling alcohol it is more soluble, but in ether it is quite insoluble, and may be precipitated from alcoholic solution by ether. It has no action on polarised light. Its melting point is from 1 60° to 165°; at 200° ib boils, and may be distilled, a portion being decomposed; at higher temperatures it car- bonises. It does not reduce cuprous oxide. All these properties readily distinguish it from the other sugars. Chemically speaking, maniiite is a hexatomic alcohol. Mannite may be separated from honey by boiling a weighed quantity of the honey with alcohol, evaporating down the alcoholic extract to dryness, and boiling this extract with absolute alcohol, concentrating the alcohol solution, and i)recipitating with ethei". Dr. Brown has published some analyses of honey, in which ihe different sugars have been identified ; the general results of six of his samples may be thus stated : — Per cent. Water (expelled at 100°), 18-07 Water expelled at a much higher temperature and loss, . 7 "99 Levulose, 36-22 Dextrose, 37 'SS Ash, -li 100 00 The chief results of twenty-five analyses made by Mr. Hehner* of honey believed to be genuine, are as follows : — Mean of the twenty- five samples. Maximum. Minimum. Per cent. Per cent. Per cent. Moisture, ISS 23-04 15-09 67-85 75-34 61-42 13-35 19-17 8-48 1-24 2-36 nil. 604 7-67 4-30 = + 1° = - 1° :: - 11° Glucose, . Difference, °° £ >, ( Glucose after fermenta- K "c 1 tioo, « § ./, 1 Total solids after ferm enta- il I "H. ( tion, In five of the twenty-five samples, polarisation In one sample, polarisation In one sample very crystalline, Li the remaiumg eighteen. Five samples which were considered adulterated yielded the following values : — 1. 2. 3. 4, 5. Moisture, 17-54 18-68 21-23 1890 21-25 Glucose, ■' 48-45 49-06 58 -.32 Difference, 34-01 31-66 20-45 Glucose after inversion, . . 43-33 48-77 10 per cent, solufion polarises . + 50° + 35° -I- 15° + 35° + 33" Glucose after fermentation, . 9-02 7-59 3-69 5-98 5-15 Total sohds after fermentation, 31-45 25-33 53-29 23-36 18-38 Difference, . . . . 22-43 17-74 4960 17-38 13-23 10 per cent, solution ])olarises alter fermentation, . . + 30° + 28° +7° -t- 16° + 10" * Anahjd, April, 1884. §76.] HONEY. 155 Dr. E. Sieben* has publislied analyses of sixty samples of lioney which he believed to be perfectly genuine — the general results are as follows : — Mean. Maximum. Minimum. Moisture, 19-98 24-95 16-28 Grape sugar, . . . . 34-71 44-71 22-23 Levulose, 39 24 49 25 32-15 Invert sugar, . . . . 70oO 79-57 C9 95 By boiling with acid cane sugar, . lOSt 8-22t 0-00 Total sugar, . .... 75-03 81-74 70-20 Dry substance, . . . .80-03 83-72 75-05 Substances other than sugar, . 5-02 8-02 1-29 The chief adulteration of honey is the addition of starch sugar in tke form of syrup. Cane sugar may be also found, and mineral adulterations are possible. Starch sugar may be detected by dialysis (0. Haenle). 200 grms. of honey are dissolved in water, made up to a litre, and the solution placed in a dialyser, so arranged that the outer vessel has a continuous stream of water running through ; at the end of fourteen hours, the solution in the dialyser is decolorised by charcoal, suitably concentrated and examined by polarised light ; natural honey treated in this way has no effect, but starch syrup turns the ray to the right. Mr. Heliner determines the moisture at 100°. The glucose is estimated by Fehling's solution — both before and after in- version, the inversion is produced by heating with 10 per cent, ■of hydrochloric acid to about 70°. The rotatory power of a 10 per cent, solution is determined both before and after fermenta- tion. J The fermentation is produced in a 10 per cent, solution by the addition of a little yeast, the vessel being kept in an in- cubator at 30° for from five to six days ; this operation might be made more speedy by fermenting in a vacuum with plenty of yeast, as suggested by Boussingault. After fermentation, the solid matter is determined and subtracted from the percentage of glucose left unfermented. The proportion of unfermentable matter should be no larger than would be yielded by a pure glucose solution after fermentation, viz., about 5 per cent. A pure honey has the following characters : — The moisture does not exceed 23 per cent. The percentage of glucose before and after inversion is about the same. The unfermentable * " Ueber die Zusammensetznng des Starkezuckersyrups, des Honigs, u. liber .<^•Tt*®'^T^T* ^o :Tfcot^ooiocj Grape sugar found after de- struction of levu- loso by hydro- chloric acid. •ji!gns eniBO eqj raojj peAu -op jTigns edujg -1- osqooBS pn-e gmiqa^ £q noijuniiiso oj gntp -JOO0T3 JTjgns ed'BJgssei JO ojoiv 6 rH +++++++ oi p. c. 33-05 44-40 43-60 51-43 52-90 56-02 55-70 •j'Bgns nvm J3TI10 soou'Bjsqtig 00 •eom;^sqns Sjq t> p c. 79-15 63-52 81-46 81-35 82-19 80-06 8188 80-27 •J0}13Ai 6. 20-85 36-48 18-54 18-65 17-81 19-94 18-12 19-73 ■* + S + T snranioo •J'BSns its^ox iri p.c. 76-84 61-67 69-00 41-35 58-50 56-04 34-75 52-30 •jugns Qu^o '* p.c. 19-45 10-62 7-06 •nonnios B.gmiqs^ ^^M. no]%v-i}i% jaij'B pono^ioaj jugns ?jOAni « .56-39 51-05 69-18 41-30 58-83 49-04 35-00 52-56 •asoinA8i cq p.c. 25-42 31-80 19-60 23-89 23-51 12-83 22-30 •jTjgns ed^JO - p.c. 25-63 37-20 21-75 34-61 25-47 21-92 30-00 § 77.] TREACLE, MOLASSES. 157 mattei* should not exceed 8 per cent. The polarising power of a 10 per cent, solution, both before and after fermentation, should be very small or nil. There should be only a slight precipitate with either alcohol or baric chloride. The ash always contains from 'Ol to 'OSS per cent, o'f PgOg. The ash from honey made from glucose larger quantities. The ash of pure honey is always alkaline, that made fi'om glucose neutral. There is an artificial honey in the market, sold as such, that consists of dextrose and levulose, and, according to Hehner, can only be distinguished from genuine honey by the entire absence of phosphoric acid from the ash.* A cariosity of food is a commercial American ai'tificial honey, which is entirely composed of glucose syrup, while the comb is also artificial, and made of paraffin. The appeai-ance of both comb and syrup is said to be superior to that of natural honey. It is not probable that this artificial honey will be met withj if there should be any suspicion that the comb is artificial, the presence or absence of paraffin is easily ascertained. Pure bees'- wax melts at 62° to 65°. Its specific gravity is -962; it contains cerotic acid, myricine, as well as ceroleine ; and, like other fatty matters, it is attacked and blackened by warm sulphuric acid. Paraffin, on the contrary, remains unacted upon, so that this test alone will suffice either to detect paraffin when pure, or to separate it from other matters, such as waxes and fats, which are carbonised by sulphuric acid. TREACLE, MOLASSES. § 77. Treacle, molasses, golden-syrup, and similar terms, are \ised to denote a sweet syrup which is produced in the manufacture of sugar, and contains a mixture of sugar, partly cane and partly fruit ; but the cane sugar, owing to certain salts and impurities, is uncrystallisable. The composition of these brown syrups varies according to the manufacture from wlaicli they are dei-ived. The cheapness of treacle, ifec, is such that there is no very great temptation to adulteration, and no conviction under the Sale of Food and Drugs Act has hitherto been obtained for adulterated molasses or treacle. The probable mode of adulterating the treacle would be by diluting with water. Cane-sugar molasses is alone used as an article of food, beet-root sugar molasses having an unpleasant taste. Some analyses, made a few yeai's ago by Dr. Wallace,! of molasses, treacle, and golden-syrup, are as follows : — * Analyst, Dec, 18S5. f "The Sugar Cane," 1869. 158 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 78. TABLE IXa. Cane sugar, Fruit sugar, ..... Extractive and colouring-matter, . Salts, Water, Specific gravity, .... W.Indian .Molasses. Treacle. Golden Syiup. Beet Sugar Per cent. 47-0 20-4 2-7 2-6 27-3 Per cent. 32-5 37-2 3-5 3-4 23-4 Per cent. 39 33-0 2-8 2 5 22-7 Per cent. 46-7 •6 15-8 13-2 23-7 100-00 100-00 100 00 100-00 1-360 1-430 1-415 r4J[)5 Tlie ash of beet molasses has the following composition : — Per cent. Potassic chloride, . . . . . . .18-70 Potassic sulphate, 4*18 Potassic carbonate, ....... 53-80 Sodic carbonate, 20-81 Calcic carbonate, ........ '35 Magnesic carbonate, "27 Moisture and loss, 1*89 JAM. § 78. Jam consists of various species of fruit preserved by boiling in strong syrup. Most jams are very readily adulterated, since any tasteless vegetable tissue, such as vegetable marrow, turnips, &c., when mixed in jam cannot be readily detected by the palate. The chemical composition of the various jams is simply the chemical composition of the fruit juice and fruit itself, with the loss of a few volatile constituents and the addition of cane sugar. The latter may be in part inverted by the action of the organic acids or ferments so constantly found in fruit. The detection of adulterations of jam is mainly micro- scopic; but, at the same time, in many cases a careful observation of the absorption-spectrum will assist the diagnosis. In order to carry out this successfully, in addition to the precautious before described, it will be safest in all cases to use comparison liquids; and those who devote themselves to this study, should have at hand a variety of genuine jams of different ages. Tlie mean composition of the more common kinds of fruits is detailed ia the following table [Kiiiiig^ : — § 79.] JAM. 159 TABLE X.-100 PARTS OF THE SEED FRUIT. 1 is II 1 ill ^ g| p 3 C) < gg b M r •a § Apple, 83-58 0-.39 0-84 7-73 5-17 1-98 0-31 Pear, . . . 83 03 0-36 0-20 8-26 3-54 4-30 0-31 Plum, 81-18 0-78 0-85 6-15 4-92 5-41 0-71 Prune, . . 84-86 0-40 1-50 3-56 4-68 4-34 0-66 Peaches, . 80-03 0-65 0-92 4-48 7-17 6-06 69 Apricots, . 81-22 49 1-16 4-69 6-35 5-27 0-82 Cherries, . 80-26 0-62 0-91 10-24 1-17 6-07 0-73 Grapes, 78-17 0-59 0-79 24-36 1-96 3-60 0-53 Strawberry, 87-66 1-07 093 6-28 0-48 2-32 0-81 Raspberry, 86-21 0-53 1-38 3 95 1-54 5-90 0-49 Bilberry, . 78-36 0-78 1-66 5-02 0-87 12 29 1-02 Blackberry, 86-41 0-51 1-19 4-44 1-76 5-21 0-48 Mulberry, . 84-71 0-36 1-86. 9-19 2-31 0-91 0-66 Gooseberry, 85-74 47 1-42 7-03 1-40 3-52 0-42 Currant, 84-77 0-51 215 6-38 0-90 4-57 0-72 Brief Notes of the Microscopical Structure of Certain Fruits. § 79. Apples and Pears. — Both apples and pears contaia numerous dotted ducts and spiral vessels. There is no very- distinctive peculiarity about these ducts, but in the core will be found a strong horny membrane with spiculated cells, crossing one another at right angles, foi-ming altogether a very singular tissue, and one which, once seen, can always be recognised. Damson. — The skin of the damson is composed of at least two distinct species of cells underlying the transparent epidermis. One kind is a double row of reddish-purple oblong or oval cells, having, when seen in section, an average length of -00232 inch, and an average breadth of about -000928 inch ; seen from above (as in tearing off a shred of the tissue) they form a beautiful five- and six-sided mosaic pattern, the size of the cells being from about -000928 to -00116 inch. The blue cells are very similar in shape and size to the reddish-purple ; below the blue there are some loose cells containing chlorophylh Hence the beautiful colour of the damson is the combined effect of the blue, the red, and the green shining through tlie triinsparent epidermis. The pulp contains the usual large colourless globes or cells, of -0116 inch average diameter (6, fig. 21). Spiral vessels are numerous 160 FOODS : THEIR COMPOSITION AND ANALYSIS. [§79. stomata are occasionally to be seen on the surface of tlie dark- coloured epidermis. The breadth or thickness of the skin is -00814 inch. By the use of bleaching powder, a small portionof theskin may be deprived of its colour, either partially or wholly, according to the judgment of the operator, and then will be seen a mapping out of the whole surface into Fig. 21.— o, Epidermis of damson ; h, lo^^es by cells so placed that pulp cells, X 115. they form a network. Flum. — There are at least three distinct structures to be seen in the boiled and preserved plum : — 1. The epidermis, consisting for the most part of a pavement-like layer of little square or irregularly oblong cells, filled with a granular matter (c, fig, 22), the size of the cells averaging from about -000696 to -00116 inch ; the general distribution of these cells is somewhat circular. ■Scattered toleraldy uniformly are patches of a deeper colour with larger cells, the patches being irregularly circular, and the centre of the patch an empty space, which possibly is a much deformed stoma. The pulp consists of the very common large globular cells {a, fig. 22), of about -12 to -14 inch diameter, almost per- fectly transparent, with a shrivelled mass within. Lastly, there are some beautiful masses of compound cells, varying in size from •016 to -48 inch (6, h, fig. 22), the length usually being from one and a half to three times the breadth. These compounds are either prismatic in shape or oval, while a few resemble long tubes. The number of cells thus bound to- gether is very vari- able, since from seven up to twenty-seven may be counted on one side. The little cellular members of the composite are Fig. 22.— Structures found in the plum, xll5. five-sided cells of an lated, and the cuticle covered with stellate hairs. ., Currants. — Both the black and the ^l red currant are similar in structure : the epidermis is covered with an ex- Fig. 24.— a, Tulp cells of cessively thin membrane, sl^owing sinu- ^-be-/';,^^' ^.^t-- ous wavy divisions, and set with simple ^ ' hairs. Beneath the outer membrane are the colour layers, con- sisting of little square masses with rounded angles about -00029 to •00039 inch diameter (a, hg. 2-5). ^^^.^^^j^ H.JSI 6 The pulp is made up of thin-walled cells, and, lastly, here and there . _ , may be found peculiar compound b^AJ^^^SsMi. >^^^^^ bodies, b, attached to the inner layer of the epidermis. These are about -0058 inch in length and -0015 inch in breadth, and are formed of .^ig. 25. -a A shred of epider- 1 f. 11 m C3„ f. mis, showing tnesmuousmarkmgs a number of oblong cells. So far i^ one portion, and the under layer as known, these bodies are found of cells in another; b, the corn- only in the currant. pound bodies, x 115. § 80.] STARCH. 163 SACCHARIN. This is the popular name given to a crystalline substance discovered by Fahlberg and Reinsen (Z)ew<. Chem. Ges. Ber., xii., 409-573) in 1879. It Avas first obtained by the oxidation of orthotolueue sulphonamide by per- manganate ; its formula is CrHjOaSN, and it has been named auhydro-ortho- sulphamine-benzoic acid CcH4 NH — or, shorter, Benzoic sulpha- mide— it is in the form of white crystals, soluble in hot water, alcohol, and ether. It melts at 220°, and can be sublimed without undergoing decom- position; it forms crystalline compounds, with the alkalies and alkaline earths; on evaporating an aqueous solution strongly acidified with HCl, almost to dryness, it is transformed into ortho-sulpho-benzoic acid. It is so intensely sweet that 1 part in 10,000 of water is very jjerceptible ; the sweetness is like that of sugar, but with a peculiar flavour. The substance is used in commerce, and the analyst will look for it in all sweet manufac- tured liquids, such as lemonades, temperance drinks, and liqueurs; it is said also to be added to sugar itself, to increase its sweetening power. It is not poisonous, but, on the other hand, it has no nutritive powers; it is said to pass through the kidneys unchanged. A general method of detecting saccharin is to shake liquids, after feebly acidifying, with ether; to separate the ether and evaporate to dryness. The ethereal extract obtained in this way, if saccharin be present, will taste extremely sweet, and if the residue is gently fused in a ])latinum dish with six times its weight of pure sodic carbonate and potassic nitrate, the sulphur will be oxidised into sulphate, and the fused mass will, when dissolved and the solution acidified with HCl, give a precipitate with baric chloride. Saccharin fused gently with potash is converted into salicylic acid, and, therefore, with ferric chloride strikes a violet colour. Solids, such as sugar, are also treated with ether, and the ethereal extract examined as before ; but if the solid has an alkaline reaction, it is best first to extract with hot water, acidify the aqueous solution, and shake out with ether. Fatty substances or liquids must first be freed from fat, by treatment with light petroleum. A quantitative estimation is most accurately made by oxidation of the sulphur by fusion and precipitating the sulphate with barium chloride — 1 part of barium sulphate equals '785 of saccharin. STARCH, CgHioOs.* § 80. It is convenient to consider the starches together, more especially as, however varied in form, the chemical composition of all starch is very similar, if not identical. Every starch corpuscle is composed of at least two probably isomeric bodies, the one " granulose," soluble in saliva, and coloured blue by iodine ; the other coloured by iodine pale * T. Pfeiffer and B. Tolleq^ {Lieb. Annalen, 285-309) have prepared com- poimds of the alkalies with the carbo-hydrates, and from the data thus obtained consider the formula of starch to be either C24H40O20 or CJ4H40O20 + HoO, and according to this formula, the amount of dextrose yielded by starch should be 108 'll per cent. 164 foods: their composition and analysis. [§ 80 yellow, and only becoming blue after the addition of sulphuric acid; it is fully soluble in ammoniacal oxide of copper, and appears to agree very closely with the characters of cellulose. These two substances may be most readily separated by diluted chromic acid, which dissolves granulose vexy easily, whilst cellu- lose remains unaltered. All starch is very hygroscopic : wheat starch, dried in a vacuum, still contains 11 per cent, of water, and air-dried from 16 to 28 per cent, of water. Starch is insoluble in cold water or spirit. Some chemists, indeed, assert that if finely powdered in agate mortars, or with quartz sand, a small portion dissolves; others contend that this is no true sokition, but the starchy matter in a state of most minute division. If warmed with water, the starch granules swell, and when heated up to 100° most starches form a semi-solution in water. True compounds of starch with bases are scarcely established. Lime and baryta appear to form weak unions, and the intense colour produced by iodine, as well as bromine, seems to point to the formation of haloid combinations. Fritsche, indeed, states that he has isolated the iodide and the bromide of stai'ch, the former containing ten equivalents of starch and one of iodine. Starch heated in closed tubes up to 100° changes gradually into soluble starch. If the temperature is raised up to 160° or 200°, it forms a transparent mass, consisting wholly of dextrin. At 220° to 280° still further change is produced, and the result is pyrodextrin, a substance easily soluble in water (but insoluble in absolute alcohol and ether), and with the composition of C^gHggOggHO. At still higher temperatures there is carbonisa- tion, and the formation of products similar to those caused by the decomposition of sugar. Starch is easily changed into sugar by the action of dilute mineral acids, as well as by oxalic acid, aqueous chloride of zinc, and by certain ferments — diastase, saliva, yeast, &c. The estimation of starch in organic bodies is always based upon converting it into glucose and estimating the glucose. This convei'sion is done in various ways. 1 to 1-3 grms. may be heated in closed tubes or flasks, in 40-50 cc. of -2 per cent, sulphuric acid for eight hours in a glycerin bath to 108° or 110°; or the substance, in some instances, may be heated in ordinary flasks in the water-bath, with a 2 per cent, solution of hydro- chloric acid, for many hours until it ceases to give a starch reaction with iodine. An excellent general method has been proposed by Dragen- dorff": — 2 to 3 grms. of the powdered and dried substance are heated, with 25 to 30 cc. of a 5 per cent, solution of potash in absolute alcohol, for from eighteen to twenty-four hours in the § 80.] STARCH. 165 •water-bath, filtered hot, through a weighed filter, which is ash free; the residue on the filter is washed first with hot absolute and then with cold ordinary alcohol, and lastly with water — the residue is now dried at 110° and weighed, the loss of weiglit corresponds to the albuminoid matters, the fat, the sugai', and the soluble salts, which have been removed by the alcoholic potash, the alcohol, and the water. The filter and its contents are now divided finely by scissors, and boiled with 5 per cent, hydrochloric acid until a blue colour is no longer struck with iodine, the liquid is then filtered through a weighed filter, and the residue washed, dried, and weighed — the dill'erence between the weights of Nos. 1 and 2 gives very nearly the starch. This weight may, of course, be controlled by estimating the glucose by Fehling. A general method for the estimation of starch in flours has been worked out by 0. O. Sullivan {Journ. Chem. Soc, Jan. 1, 1884, No. ccliv., 2-10); its principle is the freeing of the finely divided substance from fat, albuminoids, and amylans by suitable solvents, and then transforming the starch by the action of diastase into maltose and dextrin, the proportions of which are estimated by the Fehling solution and by the polariscope. 1. Preparation of the Diastase. 2 to 3 kilos, of finely-ground pale barley malt are steeped in water just sufficient to cover the whole. After standing several hours it is filtered by means of a filter press ; and if not clear by passing it also through an ordinary filter. The diastase is now precipitated by alcohol sp. gr. '83, the alcohol being added so long as the precipitate is flocculent, but discontinued when a milky or opalescent appearance commences. The diastase is washed with alcohol (•8G--88), dehydrated with absolute alcohol, and dried in vacuo over sulphuric acid. Diastase thus prepared is a white, dry, friable, soluble powder retaining its activity for a long time. Freeing the flour from fattymatters. — 5 grms. of the flour are first saturated with alcohol sp. gr. -82, and from 20 to 28 cc. of ether added. The flask containing the mixture is set aside for a few hours, and the whole is then filtered ; the residue being washed with ether. 2. Removal of Sugars, Albuminoids other than Casein, and Hatters soluble in weak Alcohol. To the flour now fat free, 80 or 90 cc. of alcohol sp. gr. -90 are 166 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 80. added, and the mixture kept at 35° to 38°, with occasional shaking for a few liours. The alcoholic solution is then passed through the same filter which has been used for the "ether" operation, and the residue washed by decantation with the same strength alcohol, and at the same temperature. 3. Solution of the Amylans. The flour which has been treated with alcohol and ether, is now submitted to the action of water ; the flour is digested with half a litre of water, and decanted through a filter at the end of twenty-four hours ; it is then repeatedly washed with water at 35° to 38°. This part of the operation is tedious, for the filtration is some- times very slow. 4. Conversion of the Starch. The residue is finally transferred to a beaker, and boiled for a few minutes in 40 or 45 cc. of water with constant stirring — it is then cooled to G2° or C3°, and -021 to -038 grm. diastase dissolved in a few cc. of water added. In a very short time the solution ceases to give a starch reaction with iodine, but it is best to maintain the digestion for an hour, because filtration is then easier. At the end of that time the contents of tlie beaker are boiled for eight or ten minutes, thrown on to a filter, and the filtrate received into a 100 cc. measuring flask. The residue is carefully washed with small quantities of boiling water at a time. When the flask is nearly full, its contents are cooled down to 15°-5, and made up to 100 cc. with water at that temperature. Should the filtrate exceed 100 cc, it is to be concentrated to the jH-oper quantity. The specific gravity of the solution is now taken ; its optical activity is determined, and its reducing power on copper solution estimated by boiling with Fehling. The optical activity of maltose Mr. Sullivan gives for the concentrations to be dealt with as [«]_,- = 154° (a^ = + 139°), and that of dextrin [a],- = + 222° (a^ = + 2004); with these values 1 grm. maltose in 100 cc. solution gives a deviation with a Soleil-Ventzke-Scheibler saccharimeter in a 200 mm. tube = 8-02 divisions, and 1 grm. of dextrin in 100 cc. in the same length of tube 11-56 divisions ; if the observations are made with any other sodium fiame polai'imeter,the factors proper to these instruments must be substituted. An example from Mr. Sullivan's paper will make the above clear. § 80.] STARCH. 167 5 grras. of barley flour, treated as above, gave as tbe ultimate result 100 cc. solution having a sp. gr. = 1-01003 = 2-539 grms. solid matter.* 9-178 grms. of this solution reduced -241 cupric oxide, and a layer of it 200 mm. in length gave a deviation with the Soleil-Ventzke-Scheibler saccharimeter = 21*1 div. From these data we have : — 0-241 X 0-7256 (K of maltose 62-5) = -1748 grm. maltose in 9-178 grras. solution. The weight of 100 cc. solution is from the gravity 101-003 grms.; hence, percentage of maltose 9*178 : 101-003: : -1748 = 1-923. Then— 1-923 X 8-02 = 15-422 optical activity of the maltose, 21-1 - 15-422 = 5 -678 optical activity of the dextrin, and 5*678 4- 11-56 = -491 the dextrin in 100 cc. solution. We have then in the 100 cc. — Maltose, . . . 1-923 grms. Dextrin, . . • '491 „ Diastase, . . . "030 „ ?444 Against . . 2-539 ,, as represented by the sp. gr. Leaving . . '095 ,, matter unaccounted for. This -095 grm. of unestimated matter was partly referred to 62 grm. a amylan, leav Mr. Sullivan's determi flour, &c., are as follows: Per cent Barlej' flour, 46-3 „ malt, 39-9 Wheat flour, 55-4 „ malt, 43-4 Rye, 44-460 Eice, 75-77 Maize 54 58 Oats, 35-38 * Taking 1-00395 as the sp. gr. for a solution containing 1 grm. of starch products in 100 cc. •062 grm. a amylan, leaving a total error of only -012 grm. Mr. Sullivan's determinations by this method of the starch in 168 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 80. A method of starch estimation devised by Marclcer and Morgen,* much used on the Continent, is as follows : — 3 grms. of the finely-powdered substance is digested in a small metal vessel with 50 cc. of water at a heat of about 100° for twenty minutes, then cooled to 70°, and 5 cc. of freshly-prepared malt extract (100 grms. malt to 500 water) added; the mixture is now kept at 70° for twenty minutes, in order to liquefy the starch paste. Then 5 cc. of a 1 per cent, solution of tartaric acid is added, the vessel covered with a metal cover and sub- mitted in a Soxhlet's digester to heat at a pressure of 3 atmo- spheres for half an hour. The digester with its contents is now allowed to cool, the lid imscrewed, the metal vessel removed, a second 5 cc. of tartaric acid added, and the whole heated once more to 70° for twenty minutes. The solution is transferred into a flask of 250 cc. capacity, filtered, diluted to 200 cc, and inverted by boiling with 15 cc. of HCl of 1*125 gravity. After three hours' boiling under an inverted condenser, the liquid is cooled, almost neutralised by soda, made up to 500 cc, and the sugar estimated by Fehling's method. A. Leclercf has proposed to treat 2 grms. of the moistened starchy substance with 180 cc. of concentrated neutral solution of zinc chloride, and to heat the mixture in a salt bath up to 108° for two hours or more, until a solution is eff'ected. On cooling, the solution is made up to a definite bulk, filtei'ed, and 25 cc. of the filtrate precipitated by 2 cc of hydrochloric acid and 75 cc. of 90 per cent, alcohol ; the precipitate is stated to consist only of starch and dextrin. The precipitate is washed with acid holding alcohol, then with 90 per cent, alcohol, dried, and weighed. It is finally burned, and any ash subtracted. M. HonigJ heats the starch-holding substance in glycerin ta 210°, and pours the solution into strong alcohol, and when quite cool adds a fifth of its volume of ether. The precipitate contains all the starch, which may be converted into sugar in the usual way. Salicylic Acid Method. — Both salicylic acid and benzoic acids dissolve starch ; and on heating the solution, the starch is con- verted into sugar. This is a good method in all cases in which the solution thus obtained can be sufiiciently clarified for obser- vation by a saccharimeter. 5"376 grms. of the starch are heated for half an hour with 100 cc. of water and 0'5 grm. of salicylic acid. The solution is clarified by adding a few drops of ammonia. * Handhuch der Spiriius-fabrikation, 1886. i Journ. Pharm. Chim., 1890. J Chem. Ztg., 1890, p. 902. §80.] 169 or sodic hydrate solution ; it is made up to 200 cc, filtered, cooled, and examined by the polariscope. Inversion by Oxalic arid Nitric Acid. — 3 grms. of the material are heated with 100 cc. of a saturated solution of oxalic acid for one hour. The liquid is cooled and made up to 200 cc. with 10 per cent, nitric acid, filtered, and the filtrate heated one hour in the water bath, the flask being connected with an upright condenser. The solution is then polarised. Inversion by Oxalic Acid. — This method has been already detailed (see p. 138). W. E. Stone,* in an elaborate study of methods for the determination of starch, gives preference to that in which the starch is converted into malt extract or diastase, for it appears that although the various processes by which starch (by means of a mineral or organic acid) is converted into dextrose are accurate with pure starches, such processes are often most inaccurate in the presence of the xylans (gum, kc), substances which, without any starch at all, give sugar indica- tions from 4 to 60 per cent. The results of his experiments are embodied in the following Table :— TABLE Xa. -GiviNo THE Results of Estimations of Starch BY Different Methods. Inversion by HGl. Inversion HNO3. Inversion by Oxalic Acid and HNO3. Solution by Salicylic Acid. Precipi- tation Barium Hydrate. Potato starch, 85-75 85-5 85-75 85-47 85-58 Dried potato, 70-92 69-79 68-53 64-25 Wheat flour, 77-G9 70-65 65-29 69-38 59-76 Cora meal, . 73-24 66-81 70-55 6211 Wheat bran, 65-86 40-25 38-68 70-77 Hay, .... 3-48 19-10 19-10 66-47 Wheat middlings, 30 00 63 09 60-24 60-44 Cotton-seed meal, 4-15 ... 54-65 Mixture of starch, sugar, and dextrin, , 9-58 2100 24-08 18-8 33-99 * Journ. Am. Chem. Soc, xvi. , No. 12; Chem. News, Ixx. 308. 170 foods: their compositiox and analysis. [§ 80, Microscopical Identification of Starches. The successful microscopical examination of starches requires practical study, and those who desire to identify them must use all drawings and descriptions as guides merely. It is not easy to preserve starches mounted as microscopical objects,* and the analyst is therefore recommended to fit up a little case, in small, wide specimen-tubes, so that he can have at hand a sample of every kind of starch possible to be obtained. These samples should be arranged in the five classes described, pp. 171-175, partly based on Dr. Muter'sf classification. A high magnifying power is not required, save for the very minute starches, such as rice and pepper. For ordinary work a magnifying power of 250 diameters is ample. Dr. Muter's classi- fication of starches was founded on observations with a B micrometer eyepiece and a ^^-inch power. It is also useful to observe the various samples of stai-ch, and make tables of their dimensions. The proper way to do this is to put the smallest possible quantity of the well-mixed starch on a glass slide, add a droplet of distilled water, cover with a thin glass, take the exact size of all the starches in the field, enu- merate them, and work them out into percentages for future reference. The illumination of starches is to be particularly attended to. The light must strike obliquely through the granules, in order to observe the rings, which are by no means so easily seen as diagrams would indicate. Polarised light is also useful, especially in the diagnosis of certain starches. Thus, the polarised starch of wheat, when examined in water, exhibits a dull ci"oss ; that of jalap, in shape and size like wheat, polarises brightly. Polarised light, in conjunction with a selenite plate, will also be found of great service. Red and green selenites are best, and give a beautiful play of colours with the arrow-roots and potato starch ; while the starches of wheat, barley, rice, and oats, scarcely show any- colour. The whole of the starches of the Leguminoste are, so far as they have been hitherto examined, likewise destitute of this power of brilliant colouration. A -|-inch object-glass, with an A eyepiece, will be found better adapted for this method of research than higher powers. If adulteration in any case has been made out, approximative quantitative results may be obtained by making a standard * According to Muter, a mounting medium of 1 part of glycerin to 2 of water preserves the characters of starch longest, i " Organic Materia Medica." London, 1S78. § 80.] STARCH. 171 niixtiu'e of the genuine starcli with the adulterant found, and then counting the individual grains in the microscopic field. Thus, for example, supposing oatmeal to he found adulterated -with barley-starch, and from a preliminary examination the mixture is thought to be 40 per cent., we proceed as follows : — Pure barley-meal and oatmeal are carefully dried at 100° and mixed so that the mixture is exactly 40 per cent. A few grains of this powder ax'e now rubbed up with glycerine and alcohol into a smooth paste, which is then further diluted to a certain bulk, a drop taken out with a glass rod, and covered with a glass, which is gently pressed down. The number of grains of barley and oat starch are now counted, and their relative pro- portion noted, and an exactly similar pi'ocess is applied to the oatmeal in question."'' If proper care be taken to i-epeat the experiments, the result is a near approximation to the truth. If photographs are taken of these mixtures, they are always at hand for reference, and much time is saved. DIVISION I. — Starches showing a Play of Colours with Polarised Light and a Selenite Plate. Class I. — The hilum and concentric rings clearly visible, all the starches oval or ovate. The group includes tous les mois, potato, arrow-root, calumba, orris-root, ginger, galangal, and turmeric. Tous les mois, or Canna arrow-root, is furnished by the Canna edulis, nat. order Marantacece. The granules vary in diameter from -0469 to -132 mm. [-0018 to -0052 inch]. "They present themselves under several forms, the smaller being granular or ovoid, the larger pyriform, whilst the largest granules are flat, oval, and pointed at their extremities. The hilum is annular, eccentric; the rings are incomplete, extremely fine, narrow, and regular. The starch dissolves easily in boiling water ; solution of potash causes the granules to swell rapidly, and gives to the hilum and lines remarkable clearness. Tous les mois can only be confused with the potato ; the size is the chief distinction. The graniiles bur.st in water at 72°., and they give a more regular cross when examined by polarised light than those of the potato. Curcuma arroio-root, which is also, called East Indian (though the arrow-root ordinarily sold as East Indian is a Maranta), is furnished by the Curcuma angustifolia. The granules are elon- *See a paper by E. L. Cleaver, F.C.S., Analyst, January 31, 1877. 172 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 80. gated triangular, or irregularly oval, flattened, and almost trans- pai-ent. The normal measurement varies from -0304 to "0609 mm. [•0012 to -00238 inch]. The hilum is eccentric, not very distinct; the concentric rings are clearly visible, and form segments of a circle. The application of heat or a solution of potash deforms the grains in a very irregular manner ; they begin to swell about 72°. Maranta arrow-root, syn. Jamaica, St. Vincent, is dei-ived from Maranta arundinacece. The granules are somewhat ovoid, flattened, and tending to a triangular shape in the larger, but the smaller may be circular. Tlie concentric layers are always visible and numerous, but not very marked. Nucleus is central, or about i eccentric — in some circular, in others linear ; from the nucleus a little slit, filled with aii*, often goes to the edge. Length of granule 0-010 to 0-070 mm., average 0-036 mm. [- -00138 inch]. Tumefaction in water begins at 76°. The specific gravity of the starch taken in petroleum or benzole is 1-504; if dried at 100°, 1-565. Natal arrow-root is probably the produce of Maranta arun- dinacece, the same plant from which Maranta itself is derived, but growing in a diflerent climate. The majority of the granules are broadly ovate, but some are occasionally circular. The dimensions are from -0327 to -0375 mm. [-00129 to -00148 inch]. The eccentricity of the hilum ranges between J^ and ^. The laminse appear under water with special clearness, and on this account granules of Natal arrow-root have been frequently mis- taken for those of the i)otato. Potato starch, syn. Potato arrow-root. — The starch derived from the potato {Solanum tuberosum). The granules vary greatly ia shape and size, some being small and circular, others large, ovate, and oyster-shaped. The hilum is annular, and the concentric rings incomplete. In the lai-ger granules the rings are numerous and distinct. The normal dimensions are -06 to -10 mm. [-0024 to -0039 inch]. The eccentricity averages i. The granules float on chloroform. Potato starch is frequently used as an adulterant of the arrow- roots. The most reliable method of examination is careful microscopic observation, but there is also a diflerent behaviour with regard to reagents, viz. : — (1.) Maranta arrow-root, mixed with twice its weight of hydrochloric acid, produces a white opaque paste, whereas potato starch treated similarly produces a paste transparent and jelly-like. (2.) Potato starch evolves a disagreeable and peculiar odour when boiled with dilute sulphuric acid, which is not the case with arrow-root. § 80.] STARCH. 173 (3.) An acrid oil may be exti-acted from the starch of the potato, but not from that of the Maranta. Ginger. — The granules are variable in shape, but characteristic. The iisual form may be described as shortly conical witli rounded angles ; the hilum and rings are very faint. Measurement about •0376 mm. [= -00148 inch]. The remaining starches belonging to this group are distin- guished as follows : — Galangal granules, skittle-shaped, with faint incomplete rings, an elongated hilum, with a normal measurement of -0342 mm. [•00135 inch]. Calumha. — The starch granules of Calumba are variable in form, most of them are pear-shaped. They have a semilunar hilum, and faint complete rings. The measurement is about •0469 mm. [•00185 inch]. Orris-root. — The starch granules are of a characteristic, elon- gated, oblong shape, with a faint hilum. Measurement -028 mm. [•00092 inch]. Turmeric has oval, oblong, conical granules, with the rings ■well marked and incomplete. Normal measurement ^0376 mm. [•00148 inch]. DIVISION II. — Starches showing no Iridescence, or SCARCELY ANY, WHEN EXAMINED BY PoLARISED LiGHT AND SeLENITES. ■Class II. — The concentric rings all hut invisible ; hilum stellate. To this group belong the starches of the bean, pea, maize, lentil, dari, and nutmeg. The nucleus of the Leguminosaj is seen usually as a long, moi'e or less stellate, air-filled black hollow. The concentric layers are recognisable if the starch is treated with chromic acid. The starch from the bean, pea, and lentil are in shape oval, oblong, and almost identical ; but the bean and pea have both a stellate hilum, whilst that of the lentil is a long depi-ession. The granules of the bean are of two kinds, large and small ; the large are fairly uniform in size, averaging ^0343 mm. [•OOISS inch] and are bean-shaped, the small ai-e nearly round ; those of the pea are variable in size, ranging from "0282 to ^0177 mm. [•00111 to -0007 inch], the smaller size predominating. The lentil granules average ^0282 mm. [-00111 inch]. The granules of the nutmeg are of small size and of chai-acteristic shape. Measure- ment not exceeding ^012 mm. [^00047 inch]. The starcli from the dari is in small elongated hexagons; average size -0188 mm. 174 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 80, [•00074 inch]. The starch from maize varies in shape irom round to polyhedral; the granules are the same size as those of the dari; the distinguishing mark is the rounded angles of the polj'gonal granules. Class III. — Starches having both the concentric rings and hilum invisible in the inajority of granules. This important class includes wheat, barley, rye, chestnut, acorn, and a variety of starches derived from medicinal plants, such as jalap, rhubarb, senega, d-c, d-c. Wheat starch is extremely variable in size, being from -0022 to -052 mm. [-00009 to -0019 inch]. The granules are circular, or nearly so, and flattened. Polarised light shows a cross, but in water the effect is not great. Barley. — The granules of barley are of fairly uniform size — viz., •0185 mm. [-00073 inch], but a few measure -07 mm. The shape of the starch is that of slightly angular circles. liye. — Rve starch is similar in shape to barley starch. The measurements are from ^0022 to •0375 mm. [-00009 to -00148 inch]. The small granules are perfectly round, and here and there cracked. Chestnut. — The starch grains vary much in form; they are round or elliptical, or three- or four-angled, with the angles rounded. In the place of a nucleus there is almost always a central hollow filled with air. The size is small and regular, being from -0022 to -022 mm. [-00009 to -0009 inch], and this regularity of size is the chief means of distinction. Acorn. — The starch granules of the acorn are almost round, or round-oval. A nucleus may be made out after treatment with chromic acid ; eccentricity ^. Normal measurement -0188 mm. [•00074 inch]. Class IY. — All the granules truncated at one end. This class includes sago, tapioca, and arum, besides several drugs — viz., the starches from belladonna, colchicum, scammony, podo2^hyllum, canella, aconite, cassia, and cinnamon. Sago. — A stai-ch obtained from the pith of certain species of palms, especially Sagus levis, and S. Eumphii. It exists in com- merce as raw; and as prejoarec^ sago ; both have oval-ovate granules, the normal measurements of which are from -0282 to -06G0 mm. [■00111 to -0026 inch]. There is a circular hilum at the convex end of the raw sago granules, and rings are faintly visible j bub § 80.] STARCH, 175 starch granules from prepared sago have a large oval or circular depression, covering nearly one-third of each gi'anule. Tapioca is a starch furnished by the lilanihot utilissima, which is more or less altered by heat, having been dried on hot plates. This causes some of the granules to swell, and thus renders indis- tinct in some cases the original structure. The starch is in groups of two to eight, or in isolated granules. When resting on its flat surface, the granule shows a little circle, and round this is a broad flat zone; but if resting on its curved surface, the granule shows contours varying from a kettle-drum to a sugar-loaf shape ; and it can then be recognised that the nucleus does not lie in the centre, but in the axis of the graniile, and always nearer to the curved than to the flat surface. A conical hollow exists under the nucleus, filled with a substance slightly refracting light. The normal measui-ement is from -0140 to -01879 mm. [-00055 to -00074 inch]. Arum starch, sometimes called arum arrowroot, has somewhat smaller grains than tapioca; they are truncated by two facets;, the hilum is eccentric. The normal measurement is about •014 mm. [-00056 inch]. Class V. — In this class all the granules are angular in form; it includes oats, tacca, rice, and pepper, as well as ipecacuanha, starch. Oat starch or meal. — The starch of the oat is mostly polyhedral^ being irregularly from three- to six-sided — -0044 to -03 mm. [-00017 to -00118 inch]. The principal starch with which it has been found adulterated is barley ; but great caution mvist be used, for oatmeal contains little round masses extremely similar to barley. Tacca arrow-root, also called Tahiti arrow-root, is extracted from the Tacca Oceanica and jnnnatifida. The granules, when viewed sideways, are muller-shaped, with truncate or dihedral bases; when seen endways they appear circular, occasionally angular or polyhedral ; sometimes a sort of contraction gives them a sub- pyriform appearance. The hilum is well developed, often starred. The normal measurement is from -0094 to -0190 mm. [-00037 to- •00075 inch]. It may be confused with maize starch, but tacca has sharp angles ; maize, rounded. nice starch. — Each individual grain is polygonal, mostly flve- or six-sided, here and there three-sided. If a high magnifying power, such as i or yV, be used, a starred hilum may be seen. The normal measurement is from -0050 to -0076 mm. [-0002 to •0003 inch]. 176 FOODS : TUEIR COMPOSITION AND ANALYSIS. [§ 81. Pepper. — The starch of pepper is in small polygonal granules, each of which, with a high magnifying power, is seen to possess -a. hilum. The normal measurement is from -0050 to -0005 mm. [•0002 to 00002 inch]. § 81. Vogel has given the following table to assist in the •diagnosis of different starches : — A. Granules, Single throughout, bounded by Rounded Surfaces. I. Nucleus central, layers concentric. {a.) For the most part round, at the side lens-shaped. Nucleus round or a radiating sHt. (1.) Large granules, 0'0396 to 0-0528 mm. [-0015 to -002 inch]— RYE STARCH. (2.) Large granules, -0352 to -0396 mm. [0013 to -0015 inch] -WHEAT STARCH. <3.) Large granules, •0264 mm. [-001 inch]-BARLEY STARCH. {h.) Egg-shaped, kidney-shaped, mostly a long, often a ragged slit ; diameter of starch, ^032 to ^079 mm. [0012 to ^003 inch]-LEGUMINOUS STARCHES. II. Nucleus eccentric, rings markedly eccentric or meniscus-shaped. (a.) Granule not flattened, or only shghtly. <1.) Nucleus mostly at the smaller end, '06 to •lO mm. [^0023 to ^0039 inch] —POTATO STARCH. (2.) Nucleus mostly at the broad end or towards the middle, ^022 to •OGO mm. [0008 to ^0023 inch]- MARANTA STARCH, (W. India arrow-root). {h.) Granule more or less markedly flattened. (1.) Many of the granules drawn out more or less at one end into a short point near the nucleus ; at the most, •OOO long [-0023 inch] — CURCUMA; at the most -132 mm. [0041 inch]— CANNA. <2.) Many lengthened into a disc, bean, or club-shaped form; nucleus near the broader end, ^044 to -075 mm. long [-0017 to ^0029]- BANANA STARCH. (3.) Many markedly kidney-shaped ; nucleus near the edge — SOUTH AMERICAN ARROW-ROOT {Sysirinchium galaxoides). (4.) Egg-shaped, one end thinning into a wedge form, placed one against the other, nucleus at the smaller end. •OS to "07 mm. [^0019 to ^0027 inch]-YAM STARCH. B. Starch Granules, Single or Compound. Single Starches with Eelation to the Little Granules they are made up of. Bounded by even, many-angled surfaces, or partly by rounded surfaces. § 81.] STARCH. 177 I. Granules throughout many-angled. (1.) With an evident nucleus, largest -0066 mm. ['00025 inch]— RICE. <2.) Without a nucleus, the largest -0088 mm. [-00034 inch]— MILLET STARCH. II. Among many angular forms also some rounded. (A.) No drum-shaped starches present, angular forms predominating. (1.) Without a nucleus, very small, -0044 mm. [-00016 inch]— OAT STARC H . <.2.) With a nucleus, -0132 to -0220 mm. [-0005 to -0008 inch]. (a.) Evident round nucleus, here and there the smaller combined, granules in variously shaped groups — BUCKWHEAT. (b.) Mostly a radiating or star-shaped fissure, none of the granules united — MAIZE. (B.) More or less numerous drum-shaped to sugar-hat shaped granules. (l.) Numerous eccentric layers. Largest granules, '0220 to -0352 mm. [-0008 to 0014 inch]— BATATA STARCH. <2.) Without concentric circles, '008 to "022 mm. [-0003 to -0008 inch]. (a.) The slit of the drum-shaped particles enlarged towards the flattened side, -008 to -022 mm. [0003 to -0008 inch] -CASSAVA STARCH. [b.) Slit wanting or not large. : ^ a & & s E < Original wheat, . 13-37 10-69 2-93 1-98 80-41 1-90 2-09 Flour No. 0, 6-b 12-56 8-38 3-06 0-83 87-26 Trace 0-47 ,, „ 1, 14-0 12-54 8-32 3-06 0-92 87-20 0-50 j> „ 2, 6-0 12-48 8-87 2-95 0-97 86-69 0-52 „ „ 3, 4-0 12-50 8-94 2-89 1-05 86-57 0-55 „ 4, 5-0 12-50 8-75 3-17 1-10 86-45 0-53 „ „ 5, 6-0 12-48 8-94 3-00 1-15 86-36 0-55 ,, „ 6, 4-0 12-39 9-38 3-00 1-17 85-87 0-02 0-56 „ „ 7, 12-0 12-35 9-82 3-06 1-28 85-01 0-09 0-74 „ „ 8, 6-0 12-41 1006 3-22 1-30 84-55 06 0-81 „ ,, 8, 5 12-40 12-56 2-72 1-91 81-52 0-08 1-21 „ „ 8, 5-0 11-72 14-34 3 00 3-51 75-90 1-02 2-23 „ „ 9, 30 10-64 15-02 2-55 4-02 74-20 1-55 2-66 Fine bran, . 16-0 11-35 13-50 3-06 4-54 63-64 8-71 6-55 Medium fine bran, 20 11-55 13-38 2-72 3-96 63-97 9-08 6-89 Coarse bran, 2-0 12-37 13-44 3-17 3-46 62-13 9-79 8-01 Gluten may be obtained by merely kneading the flour into a paste, and then washing all the starch out of the paste in a thin stream of water. For this purpose a rose with very fine holes may be fixed to a water tap, the flour made into a paste with water, the paste spread out on a rather fine hair sieve, and the streams of water made to play upon the paste, which is gently kneaded. When all the starch is washed out, the efiluent runs almost clear. As thus obtained it is, in the moist state, a yellowish-grey, very elastic, adhesive mass; and when dry, somewhat horny. It dissolves for the most part in alkaline liquids and in acetic acid. From the gluten the four bodies mentioned may be separated as follows : — 1. Gluten-casein. — The well- washed gluten is digested a few days with potash solution (for every 100 grms. of gluten about 3 to 4 grms. KHO). The clear solution is decanted from the insoluble residue, and precipitated by acetic acid in the least excess. The precipitate is exhausted successively with 60 per cent, and with 80 per cent, alcohol, then with absolute alcohol, and lastly with ether. The insoluble portion now consists of gluten-casein, which may be purified by solution of weak potash lye, precipitated by acetic acid, washed with water and alcohol, and dried in a vacuum. It forms a whitish-grey, voluminous. § 83.] WHEAT — -WHEATEN FLOUR. 185 earthy mass, soluble in dilute alkaline solutions, but insoluble in water, whether hot or cold. At 100° it soon changes into a modification insoluble in alkaline fluids. Its solution in very- dilute alkaline fluids becomes turbid on exposure to the air, and is preci2:>itated in a flocculent condition by the heavy metals. The elementary analysis of gluten-casein gives the following per- centages : — Carbon 52-9, hydrogen 7'0, nitrogen 17"1, oxygen 22-0, and sulphur 1-0. On heating gluten-casein with sulphuric acid, tyrosin, leucin, glutamine acid, and asparagic acid with ammonia, are among the products. The gluten-casein of rye seems to be similar to that of wheat iRitthausen]. 2. Gluten-fihrin. — The substances remaining in solution from 1 are gluten-fibrin, mucedin, and gliadin. The first is separated by distilling the united alcoliolic extracts to one-half, when it separates as a brownish-yellow mass. It may be purified by repeatedly dissolving in a little 60 to 70 per cent, alcohol, from which it separates on cooling. This property is, indeed, char- acteristic of gluten-fibrin. It forms a tenacious brownish-yellow mass, becoming horny on drying. It is insoluble in cold water ; boiling water partly decomposes and changes it into a modifica- tion insoluble in alcohol, acetic acid, and potash. The elementary composition of gluten-fibrin, according to Ritthausen, is carbon 54-3, hydrogen 7 -2, nitrogen 16-9, oxygen 20-6, and sulphur 1-0; this points to the formula Cg^HgglSrjQOjj^. 3. Mucedin. — The alcoholic extracts from 2. contain mucedin and gliadin, and are united and evaporated to dryness. The varnish-like residue is treated with ether to remove fat, and dis- solved in warm 60 to 70 per cent, alcohol, allowed to coo], and filtered from any gluten-fibrin still remaining. The mucedin is now precipitated in a fiocculent state by strong alcohol, and by a repetition several times of this operation is obtained pure. When fresh, it is a whitish-yellow, slimy mass ; when dry, a horny, crumbling mass. It dissolves easily in 60 to 70 per cent, alcohol, but is precipitated by 90 per cent. alcohoL The acetic acid solution is coloured a beautiful violet on the addition of sulphate of copper and potash, and slightly warming. The mucedin from rye and barley is similar to that of wheat [Ritthausenl. The elementary composition of mucedin is as follows : — Wheat. Eye. Barley. Carbon, 54-1 53-6 54-0 Hydrogen, 6-9 6-8 7*0 Nitroaen, 16-6 16-8 17-0 Oxygen, 21-5 223 21-3 Sulphur, -^ -5 -7 186 foods: their composition and analysis. [§84. 4. Gliadin. — Gliadin is obtained by evaporating the alcoholic solution left from 3. It then remains behind as a clear yellow- varnish, which is so tenacious that it may be drawn into threads. By treating it with absolute alcohol and ether, it is changed into a friable, lustreless mass. It is easily dissolved by 40 to 80 per cent, alcohol, and this solution is made milky by absolute alcohol and water, turbid by ether. By boiling it is clianged into an insoluble modification. If digested with cold water it dissolves, and the solution is opalescent and frothy, giving a precipitate with tannic acid and soda. Gliadin dissolves in dilute alkaline and alcoholic solutions and in acetic acid ; on neutralisation, or on addition of salts of the heavy metals, precipitates are formed. The elementary composition of gliadin is as follows : — Mean of two analyses. Kitthausen. U. Kreusler. Wheat Gliadin. Oat Gliadin. Carbon, 52 5 52 -8 Hydrogen, 6-8 7-6 Nitrogen, 18-4 17-7 Oxygen, 21-5 20-4 Sulphur, -8 1-7 W. Mayer has discovered a very important ratio between the total phosphoric acid in wheat and corn generally, and the total nitrogen — 1 part of phosphoric acid corresponding to 2 parts of nitrogen ; the extreme variations do not apj^ear to be more than from 1:1-83 to 1:2-19. Kitthausen, U. Kreusler, and Bote, have, however, found that in wheats very rich in nitrogen, the proportion may be — phosj)horic acid 1- to 1-31. ANALYSIS OF FLOUE. § 84. The analysis of flour should in all cases be preceded by a careful microscopical examination, combined with measurement by a micrometer in order to detect any foreign starches, &c. Blour in this country is pretty well free from organic admixture; but cases do occur, and have occurred in otlier countries, in which there has been, from carelessness in the cultivation and reaping, some one or other of the following seeds : — Melampyrum arvense [Scrophulariacea^], Lychnis githago, Lolium temulentum; or, in bad seasons, " blighted " and " ergotised " corn is ground up with good corn. The Meiam2^yrum arvense, ovjiurple cow tvheat, isanot uncommon flower in cornfields. Its structure is unlike that of wheat, and may be discovered by the microscope. A chemical test is as §84.] ANALYSIS or FLOUR. 187 follows : — Aboiit 15 grras. of the flour are made into a soft paste with acetic acid, diluted with double its volume of water. The paste is placed in a platinum dish, and the water and acid completely driven ofl:' by a gentle heat. The paste on section, should the melampyruni have been mixed with the flour, shows a colouration, violet or purple, according to the quantity. Quite lately C. Hartwich found a rye bread which was of a violet colour. The rye floiir from which it was made contained no less than 1-6 per cent, of the seeds of the melampyrum. An alcoholic extract of the flour showed an intense green colour, and sul- phuric acid gave a blue play of colours. The seeds not only of Melampyrum arvense, but also of M. cristatum, Rhinanthus hirsutus, A lector olophus major and minor, Euphrasia odontidis, and Pedicularis palustris, all give a violet colour to bread, and probably contain the same colouring-matter — rhinanthin.* The Agrostemvia, or Lychnis githago.f — The common corn-cockle of our fields is without doubt poisonous, containing a giuco- side, '^ saponin." The seeds are '^^j^, '"^N, A in sha])e not unlike a rolled up ^c:^ J ' '^ caterpillar, and the surface is / beset with regular rows of little — ' warty projections. The micro- ^ scopy of the seed is very chai'- acteristic (see tig. 27). £--■ The surface of the testa shows very large (-1 to -6 mm. dia- meter) thickened cells,forming ^. „., . ,. . ,, , r XI on the surface branching tub° ^^g- "Jt^^ '"^*'°° ""K *^' ''"'1°^ ^\ , , , ° . corn-cockle — w, the warty projections or ercles, beneath are tw-orows ot the seed, consisting of an extraordinary a regular parenchyma resting thickeuedand convoluted cuticular layer; on a thin epithelial membrane T, a loose parenchyma; e, the colourless composed of flat cells most of outer membrane enclosing the endosperm which exhibit a peculiar stria- ^^^ °^ ^'^'^ '^"'^'''' ''' '''''^'- tion; next comes the endosperm, composed of ordinary large celled parenchyma, tilled with very minute starch granules, and lastly, there is the embryo, which is in no way peculiar. There are, how- ever, bodies scattered among the endosperm with which every food-chemist should be practically acquainted ; they consist of egg- shaped, or spindle-shaped, finely granulated grains, from -02 to "l * Arcliiv der Pharmade, 217, p. 2S0. + The plant belongs to the nat. order Caryophyllacece, or clove- worts ; the flower is large and purple ; the stem dichotomous, from 2 to 3 feet high ; the calix is coriaceous, ribbed, with 5 linear lanceolate, constantly erect, patent, very long segments, styles downy, capsula 5-tootlied. 188 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 84. mm. in long diameter, and consist of masses of saponin, mucin, and stai'ch. They decompose slowly in cold water, but moi-e quickly on wai-ming and in diluted alcohol ; the starch is then set free, and exhibits the usual molecular motion. These bodies are found in other seeds belonging to the Nat. Order of Cloveworts, but are different in size. F. Hbneke {Landw. Versuchsstat, 1885, 6 Heft) has given the following as the maximum sizes of those he has observed: — Spergula -030 mm., beta -057 mm., spinacia -64 mm., agrostemma -122 mm., hence over -07 mm. points to coim-cockle. Should the microscope indicate the seeds of agrostemma, then an attempt may be made to separate saponin. At least 200 grms. of the flour are exhausted with 80 per cent, alcohol ; boiled and filtered hot ; the filtrate is now freed from alcohol by distillation, and from fatty matters by shaking up with ether; after separation of the ether, the liquid is concentrated to a small bulk and precipitated with saturated baryta water ; if saponin be present, the precipitate will be composed of baryta saponin. This may be collected on a filter, washed with baryta water, and afterwards suspended in a little water, and the baryta separated by passing carbonic anhydride through the liquid. It has also been noted that flour containing corn-cockle yields a larger percentage of oil to ether, and the ether extract has an acrid taste and is of a pronounced yellow colour. Saponin {CQ.2^bi^i8'^) ^^ ^ white amorphous powder, with a, pungent disagreeable taste ; applied to the nostrils it excites sneezing. It is insoluble in ether, in benzine, and in absolute alcohol ; it may to some extent be separated from its aqueous solutions by chloroform ; it is soluble in water, and a very characteristic feature of the solution is that it froths like a, solution of soap; solutions as weak as 1 per 1,000 exhibit this peculiarity. Saponin is precipitated by basic acetate of lead and in concentrated solution, as above detailed, by baryta water. Saponin is Isevogyrate [«]fl= - 7°'30. Concentrated sulphuric acid dissolves saponin with first a red-yellow colour, changing into violet, and later into a more intense red. According to Schiaparelli * saponin has a remarkable power of dissolving insoluble salts —for example, a boiling aqueous solution will dissolve barium carbonate up to 10 per cent. 2 grms. given to an adult will cause physiological symptoms, mainly consisting of nausea, diaphoresis, and diuresis; '5 grm. is a fatal dose for a kitten; the fatal dose for a human adult is not known. The Lolium temtdentum, or Darnel, has been found as an im- purity in fiour. Darnel cannot be identified microscopically by the starch granules, for they are not sufficiently characteristic; should Lowever any of the husk be found, it may be readily distinguished • Gazetta xiii,, 422-430. § 84.] ANALYSIS OF FLOUR. 189 from wheat, but possesses some similarity with that of the oat. The best distinction is that the darnel bran has laucet-shaped hairs, while those of the oat are more in shape like the upper bill of the vulture and birds of that class. If flour contain a considerable quantity of darnel the characters of the alcoholic extract assist the diagnosis. The flour is digested in alcohol of 35°; if it is pure, the alcohol remains perfectly clear and limpid, or at the most, takes a very pale-straw colour, from dissolving a little colouring-matter in the envelopes of the wheat, which may be in the flour, nor is the taste disagreeable. If, on the contraiy, it contains darnel, the alcohol takes a greenish hue, which darkens gi-adually. The taste of the alcohol is acrid and nauseous, and on evaporation it leaves a yellowish-green resin. Detection of Ergot in Flour. — It is most important to examine flour for grain matters damaged by mould, and especially ergot. A good preliminary test is that recommended by A. Yogel.* The flour is stained with aniline violet, and then examined micro- scopically: the damaged starch granules take up the colour intensely. This staining will take place with flour damaged by any fungus, and is not a special test for ergot. The best chemical method is that of Jacopy, as modified by J. Petri.f 20 grms. of the flour are placed in a proper exhausting apparatus, such as is described at page 68, and exhausted with boiling alcohol until the last alcohol is colourless. To the alcoholic solution 20 drops of cold diluted sulphuric acid are added, and the liquid is filtered and examined by the spectroscope in thinner or thicker layei's, according to the depth of colour. If the flour is ergotised, the alcoholic solution will be more or less red, and show two absorp- tion-bands in very dilute solution, one lying in the green near E, and a broader and stronger band in the blue between F and G. On mixing the original solution with twice its volume of water, and shaking successive portions of this liquid with ether, amyl- alcohol, benzine, and chlorofoi-m, the red colour, if derived from ergot, will impart its colour to each and all of these solvents. Other tests have been proposed from time to time; as, for example, a yellow colour developed when flour is treated with an alkaline solution, and the development of a smell of trimethylamine when the potash solution is heated. It may, however, be remarked that the yellow colour would not be conclusive as to the presence of ergot, for otherwise damaged flour will show this reaction; and as for the smell of trimethylamine, it may be noticed when certain gummy matters are decomposing, and though a flour pro- ducing such an odour with potash cannot be considered healthy, * Chem. Centralblatt. [3 f.] 10. 559. f Zeitschrift fur Anal. Chem. 1879, pp. 211-220. 190 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 84. the odour in itself would not be conclusive in regard to the presence of ergot. The presence of ergot, and of tlie seeds mentioned, must be regarded, not as adulterations in the sense of a fraudulent addition, but as impurities. Nevertheless, an analyst would naturally certify, vmder the " Sale of Food and Drugs Act," if flour sold as good flour should be found to contain any of these substances. The following substances have been fraiidulently added to wheaten flour : — Rye, rice-meal, barley-meal, potato starch, the flour from various Leguminosse, linseed-meal, buckwheat, and some other starches. It may once again be said, that in England all these adulterations of flour are of extreme rarity (with, perhaps, the exception of potato flour and ground rice); but there is good evidence that in times of scarcity, with bread at famine prices, all kinds of substances have been mixed with flour and sold. What has happened may occur again, and it is therefore well to know the chief chemical tests which have been recommended to detect even these uncommon admixtures. The general test recommended by A. E. Vogel may be iiseful : The suspected flour is exti'acted with 70 per cent, alcohol, to which, hydrochloric acid has been added, in the proportion of 5 per cent, of the alcohol employed. If the flour is made of either pure wheat or rye, the alcohol remains colourless ; it is of a pale yellow if either barley or oats should be present; orange-yellow with pea flour; purple-red with mildewed wheat; and blood-red with ergotised wheat. Potato Starch. — So long ago as 1847, M. Donne proposed an excellent test for potato starch in wheat flour. The flour is examined in a very thin layer under the microscope, in the ordinary way, and then, while it is under observation, a weak solution of potash is added, when potato starch will begin to swell, and reach four or five times its volume, while wheat starch is scarcely afi"ected. The test is best applied by putting a little of the flour on a stage micrometer; it is then easier ta appreciate the alteration in size of any particular starch. When this method of detecting potato starch is combined with the subsidence process, proposed by Lecanu in 1849, so small a quantity as one part of potato starch in a thousand of wheat flour may be detected. The subsidence process is asfollows : Any convenient quantity of flour, say 100 grms., is treated with 40 per cent, of its weight of water, and the gluten separated in the usual way; the washing water is well stirred, and passed through a sieve to retain the larger suspended matters, and then allowed to rest in a conical vessel until a deposit has formed. Without waiting for the supernatant water to become clear, it is decanted, and the § 84. J ANALYSIS OF FLOUR. 191 deposit mixed by stirring with more water, and allowed again ta deposit for a short time. The water is decanted, and the process again gone through. By this means the final and lowest deposit will consist almost entirely of potato starch, which, being of greater specific gi'avity than wheat, always subsides first. M. Kobine, curiously enough, relies more upon a chemical identifi- cation of the apex of the cone of deposit than upon a microscopical, which latter is so much more decisive. The last deposit is recommended to be received on a lump of dry plaster, the apex cut off" and triturated in an agate mortar — glass, porcelain, and Wedgwood mortars do not answer — and tested with iodine, which gives a blue colour with potato starch; but under these circumstances, not with wheat starch, the friction of the smooth agate not having been sufficient to tear the envelopes off" the latter. M. Chevallier has also recommended a method for the detection of potato starch, based on the resistance which the wheat granules possess to the destruction of the outer membrane. Equal weights of flour and sand are to be triturated with water until a homogeneous paste is formed, which is then diluted and filtered ; to the filtrate is added a freshly-prepared solution of iodine, made by digesting for about ten minutes 3 grms. of iodine in GO cc. of water, and then decanting. If the flour is pure, this addition will give a pink colour, gradually disappearing ; whilst if potato starch should be present, the colour is of a dark purple, only disappearing gradually ; by comparing the reaction with flour known to be pure, this difference of behaviour is readily appreciated. Detection of Legtcminous Starches, d'c. — As previously stated, the leguminous starches give no play of colours when examined by polarised light and a selenite plate, and are thus easily de- tected among the iridescent wheat starches. By treating the flour also under the microscope with a solution of from 10 to 12 per cent, of potash, it is possible to dissolve the starch granules of the leguminous plants, and leave a characteristic reticular tissue, made up, for the most part, of irregular hexagons. The addition of lentils or vetches, on account of the brown colour of the seeds, can only take place in minute quantity, and then could only be added to dark flours of inferior quality. Bean flour, haricot flour, or pea flour, may be mixed up to 5 per cent, without imparting any particular appearance, odovii', or taste. Beyond that, all these characters ai-e altered. Bean flour is said to give to the crust a more golden brown, which is agree- able to the eye. There is a principle in beans and vetches which, treated with nitric acid and ammonia, gives a red colour. The method to separate this colouring-matter from falsified flour is 192 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 84. to exhaust any convenient quantity with boiling alcohol, and to evaporate the alcohol to a syrup. This syrupy is freed from fatty matter's by ether, and the insoluble residue exposed successively to the vapours of nitric acid and ammonia. An amaranth red colour denotes the presence of these substances. M. Biot has, however, stated that wheat from the Caucasus responds to this test although perfectly pure, so that, like many other reactions, it must not in itself be taken as conclusive. M. Marten proposed to separate legumin, and M. J. Lemenant des Cheuais has modified Marten's original process as follows : — The gluten is separated in the usual manner, and to the liquid containing the starchy matters is added ammonia, which is a good solvent of legumin. The starch is allowed to deposit, the liquid is filtered, and to the filtrate a very dilute mineral acid is added, which, precipitates legumin if joresent. The legumin is filtered, collected, dried, and weighed. According to M. Lemenant des Chenais, •9 of legumin in 100 grms. of flour represents an adulteration of 5 jier cent. The most scientific process, which embraces a fiiii-ly complete •examination of flour for the leguminous constituents, is that of Lecanu : — The gluten is first separated in the usual wa3^ The washing water, containing starch, soluble matters, and legumin, if present, is passed through a sieve to separate coarse particles in suspension, and then diluted sufiiciently and allowed to deposit. The liquid is divided into two parts, and one part is allowed to putrefy or ferment spontaneously. With pure flours the lactic acid fermentation is most common ; with flours con- taining legumin there is a putrid fermentation. The other portion is, after decantation and filtration, concentrated until a yellowish scum forms on the surface ; it is allowed to cool, and separated from the albumen which all flours contain. Then legumin is precipitated by adding drop by drop acetic acid. Legumin is identified by its properties. It is without colour, taste, or odour. When dried it is of a horny consistence, in- soluble in alcohol, not coloured by iodine, but very soluble in potash or ammonia water, from which solution it may be precipi- tated by the addition of an acid. The deposit is submitted to a careful microscopical examination, and tested with iodine to colour the starch and leave imcoloured the cellular tissue, or with potash in the way described on page 191. The suspended particles on the sieve are also examined microscopically, because they often contain large fragments of leguminous cellular tissue. The leguminous starches contain more mineral matter than wheat flour — for example, pea flour contains, on an average, 2-65 per cent, of ash j flour, -7. It hence follows that if pea flour be § 85.] ANALYSIS OF FLOUR. 193 mixed witli wheat flour in the proportiou of 10 per cent., the ash would be -87 instead of -7, and it has been proposed to make this a test of the presence of such foreign starches , but, as the example just given shows, with moderate adulteration it would not be at all conclusive, and must only be considered one of the auxiliary means. M. Rodriguez has ascertained that when pure flour is submitted to dry distillation in a stone retort, and the distillate is collected in a vessel containing water, the latter will remain perfectly neutral. But if bean, pulse, or pea meal has been added, the water will have an alkaline reaction. This test appears of doubt- ful value ; for, provided the distillate is alkaline, the alkalinity may, it is evident, have arisen from a variety of causes besides the addition of the substances mentioned. It has also been shown by Bussy that certain cereals yield on distillation an acid product. Lassaigne (taking advantage of the fact that haricot beans, as well as beans, contain a tannin in their envelopes) adds a salt of iron, which, with pure flour, gives a feeble straw colour, but mixed with either of the two mentioned, or, of course, with any substance containing tannin, gives various shades, from orange- yellow to very dark green. § 85. Detection of Alum and Mineral Matters generally in Flour. — The most important test for the detection of mineral substances generally in flour is, without doubt, what is known as the "chlo- roform " test — a test which, it would a^^pear, was first proposed by M. Cailletet, a pharmacist of Charleville, in 1869, and was in England brought prominently before the notice of analysts by the researches of Dr. Dupr^. The principle of the method is simple and obvious. The chloroform is of sufiicient gravity to float the starchy substances and allow the alum, sand, sulphate of lime, or other mineral matters, to sink to the bottom. It, besides, has no very appreciable solvent action on alum, and none at all on the generality of mineral or saline substances. No solution made of sufiicient specific gravity, by dissolving salts in water, or any other means, will answer the same purpose as chloroform, because, directly the flour is moistened with water, most of the alum is decomposed by the phosphate of potash present in the flour, and alum also forms an insoluble compound with the gluten. The method is as follows : — The tube figured in the article on "Beer" is taken, and a weighed quantity of the flour, from a quarter to half a pound, is placed in it, and sufficient methylated chloroform added to form a thin sort of paste; the U 194 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 85. cylinder is closed by a stopper, shaken iip once or twice, and allowed to stand over night. The next morning the "rod- stopper" is inserted and the cap removed; the latter will contain sand from the millstones, sulphate of lime, alum, or any other mineral powder of a greater specific gravity than chloroform, that happened to be in the flour; this fluid is placed in a burette, some more chloroform is added, and the matters allowed again to subside; lastly, the powder, with a little of the chloroform, is drawn ofl" into a watch-glass, the chloroform evaporated, and the powder digested in warm water, filtered into a clean watch-glass, and allowed to evaporate spontaneously. If alum were present crystals will be obtained, easily identified by their form, and these, if necessary, can be produced in court as a "corpus delicti.'" Most of the chloroform used may be recovere'd by filtration, and purified by distillation.* The alum crystals may be easily identified by theix- form under the microscope, and by the reaction with gelatin and log- Tvood. It may be a matter of some importance to be able to say whether tlie alum present is a potash or ammonia alum. The best method of .detecting this is to take the smallest crystal, and having previously dropped a single drop of Nessler solution on a porcelain slab, stir the crystal into the Nessler ; an imme- diate brown colour and precipitate is produced if the alum was an ammonia alum. Dr. Dupre has made some experiments as to the amount of alum which by this process it is possible to recover. Three mixtures were made, containing respectively 28, 10, and 3 grains of very finely powdered ammonia alum in 100 grains of a- pure fiour. On separation of the alum by the chloro- form test, tlie residue or deposit obtained from the chloroform was dissolved in cold water, and precipitated by baric chloride, and the sulphate of baryta obtained calculated into ammonia alum; the result was that 27-1. 8-21, and 1-14 grains of alum were respectively recovered, instead of 28, 10, and 2 grains, which must be considered as fairly satisfactory. The sand and silica obtained by the chloroform process will be filtered off", and should be dried and weighed, more especially since there has been found to be a relationship between the silica present and the alumina in flour not existing as alum, but as clay, &c. The Logwood Test. — A freshly-prepared tincture of logwood be- comes blue when alum and certain other salts ai'e added to it ; an excellent and readily applied test has been proposed based on this reaction. The process usually adopted for flour is as follows: — Fifty grms. of flour are weighed out and mixed by the aid of * Emmerling, instead of cWoroform, uses a solution of zinc sulphate, 100 grms. of zinc sulphate dissolved in 72'5 grms. of water j such a solution has a gravity of 1-43 {,ZeU. f. anal. Chem., 1894, p. 46). § 85.] ANALYSIS OF FLOUR. 195 a glass rod with 50 cc. of distilled water ; to this is added 5 cc. of recently prepared logwood solution, alkalised by 5 cc. of solu- tion of ammonium carbonate. If Yohro V^^^ °^ alum is present, the flour will become of a lavender-blue colour instead of pink. An approximate estimate of the quantity may be obtained by having a standard solution of pure alum, 1 grm. to the litre, and adding known quantities to exactly similar emulsions of pure flour, and testing as befox'e with logwood, until an emvilsion is obtained of very similar hue to the flour originally tested. If the cold extract gives a blue tint with the logwood test, or if the flour be submitted to dialysis, and the difl'usate responds, alum is present as alum, and is not derived from dirt, clay, or from the millstones themselves.* The author uses little strips of gelatin to coucenti-ate the alum on : a bit of gelatin is soaked in the cold extract of the suspected flour for twelve hours, it is then taken out and steeped in the ammoniacal logwood; if alum is present the gelatin becomes of a beautiful blue colour ; and the spectrum shows the shifting of the band as described on p. 98. The same blue colour is produced by the presence of magnesia, and clayey matters may also cause a bluish tint. Nevertheless, if a flour or bread does not respond to this test, it is certain that alum in any quantity is not present; on the other hand, if a blue colour is produced, there is likely to be either an adulteration with alum or some other admixture, and the sample should be more thoroughly examined. Hermann W. Vogel t has shown that alum and magnesia salts can be recognised by their influence on the spectrum of purpurine. It is evident that here is a process by which the analyst may be assisted in his diagnosis of the cause of any bhie colour imparted to flour. Pure purpurine gives, in saturated solutions, a spectrum extinguishing wholly the blue part. An alcoholic solution diluted until it is of a straw-yellow colour extinguishes the blue only partially, and shows two marked absorption-bands at F and b E (see fig. 16, p. 94). A diluted watery solution does not show these absorption-bands, but instead there appears a stronger ab- sorption in the green between F and b, a weaker in the yellow from E. This reaction is dependent on a trace of alkali, for it is intensified by ammonia, whilst a slight excess of acetic acid colours the fluid yellow, and then there is only a weak absorption. The * The millstones are sometimes mended with an alum cement. This circumstance will of course, from time to time, be utilised for purposes of defence. + Ueber eine empfindlkhe Spectral Analytische Reaction auf Thonerde und Magnesia, 196 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 86. solution of purpurine should be prepared from purpurine which, has been purified by sublimation, and it should be made very feebly alkaline. To test for alum it is best to take the deposit from the chloroform process already described, and dissolve it in not more than 1 cc. of water. 2 cc. of water are now placed either in a test-tube or a little glass cell, and three drops of a saturated alcoholic solution of purpurine added, and then alkalised by a drop of fourfold diluted ammonia water. On observing this solution by the spectroscope, it appears as curve No. 12, fig. 16, p. 94. A drop of the alum solution is next added : in dilute solutions two bands gradually appear; in the pi-esence of half a milligramme of alum, the bands appear after the lapse of several minutes. Magnesia presents similar appearances, but is at once distinguished from alum by the fact that the bands are destroyed by the addition of acetic acid. Proximate Analysis of Flour. § 86. The constituents of flour to be determined are — (1.) Water. (2.) Fat. ! Sugar, Gum, and Dextrin Vegetable Albumen. Phosphate of Potash. (4.) Gluten. (5.) Ash. (1.) The water is taken in the ordinary way; that is, by weighing carefully about 1 to 3 grms. in a tax-ed dish, and expos- ing it to the heat of the water-bath until it ceases to lose weight. (2.) The fat, according to the researches of Peligot, must be determined in the inrfectly dry flour, error resulting in any other case. (3.) The cold extract is obtained by digesting 10 grms. of flour in 500 cc. of water, and filtering and evaporating down 250 cc. in a platinixm dish. According to Wanklyn, 100 grms. of flour yield to water — Grms. Sugar, gum, and dextrin, 3 '33 Vegetable albumen, . . . . . . 0'92 Phosphate of j)otash, 0'44 4 -69 On igniting the extract, the ash should consist entirely of phosphate of potash. "When the weight of the ash is known, it may be dissolved in water, and the quantity of phosj)horic acid § 86.] ANALYSIS OF FLOUR. 197 estimated by titration with uranium solution ; and if from this there is any discrepancy between the calculated phosphate of potash and that found, the ash should be carefully examined. The determination of the sugar and dextrin may be made by the processes described at p. 137 et seq.; but it is usually sufficient to obtain merely the weight of the cold extract and the weight of its ash. A method of estimating the value of flour by the amount of solid matter dissolved by acetic acid has been proposed by M. Eobine, who has taken advantage of the property which acetic acid, when properly diluted, has of dissolving the gluten and albumen, and leaving intact the starchy matters. The acetic acid solution increases in density according to the amount of solid substances it dissolves, and he has constructed an areometer, graduated in such a manner that each degree represents the value of the flour expressed in a loaf of 2 kilometres weight. A table is sold with the instrument, and without doubt, although not exact enough for the food-analyst, the process is of some value to the buyer of flour. The areometer is called " Appre- ciateicr des Farines." The acetic acid is diluted until the "■ appre- ciateur" sinks to 93° on the scale. 24 grms. of flour of the first quality are taken for the assay, but if the flour is of the second quality, then 32 are taken. This quantity of flour is washed successively with six quantities of the acid, each time using 31 '25 cc, and all the time triturating in a mortar. After ten minutes the whole is poured into a vessel, plunged in cold water of exactly 15°, and allowed to remain at i-est for an hour; the liquid is then decanted, and the " appreciateur " floated in it. By the number indicated, the number of loaves of bread 2 kilo- grammes in weight which 150 kilogrammes of the flour will give is at once seen. (4.) The gluten or albuminoids can only be approximately determined by the washing process desciibed at page 184; the usual method is to make a careful combustion of a small weighed portion of the flour with cupric oxide, and to measure the nitrogen obtained, and then to multiply the percentage of nitrogen found by the factor, 6 "33. (5.) The ash is burnt in the usual way, and is somewhat difficult to consume, especially if any quantity of flour is taken. It has been proposed to mix the flour with nitrate of ammonia, then to heat carefully, and directly fusion commences, to with- draw the flame. Flour can certainly be burnt up very quickly in this way. If this method should be adopted, it will be very necessary for the analyst to ignite a corresponding quantity of nitrate of ammonia in a platinum dish, and see whether any 198 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 87. residue is left. Occasionally, nitrate of ammonia may be met •with which is sufficiently impure to cause an error in analysis. Flour may also be burnt up in a platinum trough in a combustion tube. In this case it is most convenient to begin, the combus- tion in ordinary air, and then to finish in oxygen. A properly burnt flour ash should be below 1 per cent. ; if it attains 1 per cent., mineral adulteration is probably present, (The method of estimating alumina and silica in the ash of flour is fully detailed at pages 205-207, and also the relationship between the silica and the alumina.) Legal Case Relative to Flour. § 87. The following brief abstract will show the lines of defence likely to be adopted : — In the month of February, 18S0, the case of a miller summoned for selling adulterated flour was heard at the Eckington sessions. The analyst deposed to having found alum, in the proportion of 24 grains to 4 pounds of floiu-. He obtained the alum as alum by the chloroform process. He shook the flour with chloroform, which was a heavy liquid, the flour floated, and the alum sank to the bottom; it was from what sank that he obtained crystals ia the characteristic form of alum; he tasted it, and it had the astringent taste of alum. It gave the logwood reaction such as alum gives. He placed about 30 grains of the flour in the chloroform, and the precipi- tate was probably one-eighth of a grain. He let the chloroform evaporate, and so obtained the crystals ; alum crystallises in octohedra of the cubical system ; the alum was in the fragmentary form until water was added to the deposit from the chloroform, and the liquid Altered and evaporated. Silica crystallised in hexagonal prisms, and could not be mistaken for alum, besides, it was insoluble in water. He had made an analysis for the purpose of estimating the quantity of alumina present, and found it was in the proportion corresponding to 30 grains of alum to 4 pounds of flour. On being asked whether clay and dirt might not account for the alumina, the ansvver was that clay and dirt might be present as a silicate of alumina, but it would be insoluble in water, and would not give the reaction with logwood. The defence was — 1. That the analyst was mistaken. 2. That alum was occasionally used in the mill for filUng up the cracks in the stones. 3. That the defendant had made his flour lately from foreign grain on account of the bad quality of English wheat at the time, and there was nothing astonishing in flnding 24 grains of alum in such wheat, although perfectly pure and unadulterated. An analyst was called for the defence who did not seem to be acquainted with the chloroform test, but had estimated the total alumina. The gist of his evidence was that he could not say positively whether there was alum or not in the flour, and that he thought that so small a quantity of alum as could be separated from 30 grains could not be identifled. The matter was then referred to Somerset "House, and the Government chemists fully confirmed the presence of alum in the flour.* * Analyst, 1880, p. 72-8G. § 88.1 BREAD. 199 § 88. The term Bread lias been applied to any form of flour made into bread, but that made from wheaten flour can alone be treated of here. Wheaten bread is the flour of wheat made into a paste with water, and the paste is permeated by carbon dioxide, either by adding yeast, which causes a true fermentation with the production of alcohol and carbon dioxide, or the carbon dioxide is added in solution in water to the paste, as in Dauglish's system. The explanation of the bread-making process is not thoroughly worked out in all its details, but the following theory agrees fairly well with what is witnessed. On adding yeast to the dough, it is placed on one side, at a suitable temperature, and allowed to rise, that is, fermentation proceeds, and there is a con- tinual evolution of gas; the starch in some degree becomes changed into sugar, which sugar is decomposed into carbon dioxide and alcohol. The gluten prevents, or rather retards, the escape of the carbon dioxide, and the tension of the warm gas expands little cells, and gives to the bread its familiar light spongy appearance. The alcohol mostly escapes, and although in large iDread-making establishments it would seem to be feasible and economical to recover the alcohol, hitherto no i-eally good appliance has been invented for this i)urpose, the apparatuses which have been tried interfering with the baking of good bread. The outside of the loaf, when placed in the oven, is liaised to a temperature of from 210° to 212°, but the crumb is seldom much above 100°. The ci'ust is to some extent caramelised, and, on analysis, shows, as might be expected, very much less water than the crumb. Thus, Eivot found in twenty-one samples of bread from 2045 to 47-11 per cent, hygroscopic water in the crumb, and 1640 to 2744 per cent, in the crust. Tracing one by one the chief chemical changes which the flour undergoes under the influence of the yeast-fermentation and subsequent baking, we consider, 1. Nitrogenous Matters. — The soluble albumen becomes in- soluble, and can no longer be separated from the starch. The gluten-casein and gluten-tibrin form some intimate combination with the starch. The gliadin, however, still may be extracted out of the bread, as out of the flour, by the action of alcohol. In the crust there is a partial destruction of the nitrogenous substance. Thus, Y. Bibra found — Wheaten Bread. Kye Bread. Nitrogen per cent. Nitrogen per cent. Crumb, .... 1498 1-476 Crust, ..... 1-363 1-293 200 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 8Si 2. The Starch, as already explained, is in part changed into sugar, ■which svigar is further decomposed into carbon dioxide and alcohol ; a part of the alcohol appearing as acetic acid, and a part of the sugar appearing as lactic acid. There are also other volatile acids of the acetic series formed (notably with dirty breads), and small quantities of butyric acid can be obtained and identified if the watery extract (which, by the way, always reacts acid) is distilled after the manner of Duclaux. All the sugar formed is not decomposed, but the bread invariably con- tains more sugar than the flour from, which it was made. A portion of the starch is also changed into dextrin, and through all these causes the bread contains always more soluble carbo- hydrates than the flour. 3. The Fatty Matters are not, so far as is known, changed. 4. The Ash is not changed, save by the minute proportion of yeast ash which is added to it, an addition quite inappreciable. Further, any salt added by the baker increases a little its weight; but the ordinary method of burning bread volatilises very effec- tually chlorides of the alkalies, so that tlie ash of bread is still very small. It has been said that the alcohol escapes, which is true with regard to the bulk of the alcohol. Alcohol, however, has a wonderful property of adhering to organic substances, and Th. Bolas has shown that it can be detected in fresh bread in greater quantities than would a priori have been suspected. Thus, he found in six fresh samples of bread the following per- centages of alcohol : ' Minimum, ■221 Maximum, ........ "399 Mean, ........ '313 In two of the samples a week old, he was able to detect 'ISS to -120 per cent, respectively. On keeping bread, there is a continual loss of water, and it becomes " stale " from some peculiar molecular change. That this staleness is not due to the loss of water is proved by the simple experiment of re-baking a loaf, when it becomes for the time fresh again, but more rapidly after this process becomes stale and is notably drier. V. Bibra found that a bread cannot be made fresh again if it has lost 30 per cent, of watei', but if the loss of water is below that, it then may be freshened by re-baking. V. Bibra found that wheaten bread lost the following percentages of water : — After 1 day. 3 days. 7 days. 15 days. 30 days. 7-71 S-S6 14-05 17-84 18-48 The mean composition of wheaten bread, from a large number of analyses collected by Konig, is as follows : — '•J BREAD. 20 Mean for Mean for Minimum. Maximum. Fine bread. Coarse bread. Water, . . . . 26-39 47-90 38-51 41-02 Nitrocjenous substances, 4-81 8-69 6-82 6-23 Fat, " . . . . •10 1-00 •77 •22 Sugar, . . . , •82 4-47 2-37 2-13 Carbo-hydrates, 38-93 62-98 49-97 48-69 Woody tibre, -33 •90 •38 -62 Ash, . . . . •84 1-40 1^18 109 The ash of a properly burnt -vvlieaten-flour loaf seldom exceeds 1"5 per cent, unless adulterated; anything beyond 3 per cent, would be certainly suspicious of a mineral addition. There has recently been an agitation on behalf of "whole meal bread," and analyses of the greater richness of such bread in azotised con- stituents are frequently quoted ; but such a question caimot be decided by chemical analysis, or, at all events, by ordinary analysis, in which a few constituents are alone estimated. The question is rather a physiologico-chemical inquiry, and the proper way to solve the problem is to go on the lines of the well-known experiments of G. Meyer. A healthy individual is taken and fed on known weights of the substance experimented upon, and the amount of undigested substance recovered from the fpeces is weighed. Meyer thus experimented on — (1.) Horsford-Liebig bread, which is made without the addition of yeast or leaven, the carbon dioxide being developed by the action of bicarbonate of soda on phosphate of potash. (2.) Munich rye bread, prepared from rye bread and coarse wheat meal and leaven, (3.) White wheaten bread. (4.) North German black bread {Pumpernichel) prepared out of whole rye meal, and with the use of leaven. The amount of diy substance, ifec, absorbed in percentages of these different breads, was foiind to be as follows : — - Dry Substance. Nitrogen. Ash. 1. The Black Bread, . . . 80-7 57-7 3 4 2. Horsford-Liebig Bread, . . 88-5 67-6 619 3. Rye Bread, . . . . 89 9 77 8 69^5 4. White Bread, .... 94-4 80-1 69^8 It is thus shown that of the black bread a person would have- to eat very much more than of white bread. The white wheaten bread was nearly all absorbed. That this experiment was not made with whole wheaten meal is true, but it still unmistakably casts some doubt on the question as to whether whole meal would be more nourishing than pure white flour. Alterations of Bread by Moulds, d-c. — Red, green, orange, and black spots occasionally appear on bread, and there are several 202 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 89. instances on record of great damage and loss from sucli parasitic diseases. In 1856, in France, M. Poggialo was commissioned to examine 22,000 rations served out to the French troops, the bread of which had tui-ned a bluish-black. Tlie bread had been made of inferior grain, but it also contained an enormous number of bacteria. Rather frequently, also, bread and other foods have been attacked by an orange-red growth, which has been attri- buted to a fungus, to which has been given the popular name of the red-bread fungus, its scientific appellation being Oidium aurantiacum. A red algse sometimes appears on bread ; it has been named the Palmella iwodigiosa, and has been specially studied by Dr. Antoine Franchini, of Bologna. To the eye, the algse resembles almost exactly drops of blood. It is composed of cells and filaments filled with a bright red colouring-matter, which would well repay examination. The more common moulds of bread are the whitish Mucor tnucedo, the green Aspergillus glaucus, and the black Rhizopus nigricans. It has not yet been established that any of the moulds or growths envxmerated are in themselves injurious to health; but, as may be expected, they damage the bread, making it deficient in nourishment, and unpalatable. § 89. Adulterations of Bread. — The adulterations of bread enumerated by writers are sufiiciently numerous, but those actually proved to exist are but few. Among organic additions rice fiour, potatoes, bean flovii-, and pea flour are usually given ; among mineral, alum, borax, sulphate of copper, sulphate of zinc, chalk, and carbonate of magnesia. In 1843 and 1847, some bakers in Belgium were convicted of adding sulphate of copper to their bread, and this fraud has been repeated a few years ago by a baker of Calais. There is, however, no reason to believe that English bakers are addicted to these practices, and, as a fact, no conviction has been obtained save for the use of alum. The detection of rice flour, bean flour, foreign seeds is to be undertaken in the same way as described in the sections on flour, save that here the chemical tests are more useful than the microscopical. It is an extremely diflicult thing to ■detect and identify most starches when they have been swollen by heat and altered by fermentation. The only feasible course appears to be to make bread of flour adulterated with the sub- stance suspected to be present, and examine sections and wash- ings of such bread side by side with similar sections and washings of the suspected bread.* * An exception may, perhaps, be made to this statement in the case of potato starch, which may be recognised tolerably easily even in bread. § 90.] BREAD. 203 § 90. Alum in Bread."^ — Alum is added to bad or slightly damaged flour by both the miller and the baker. Its action, according to Liebig, is to render insoluble gluten which has been made soluble by acetic or lactic acids developed in damp flour, and it hence stops the undue conversion of starch into dextrin or sugar. It will be found that generally the medical profession believes that aluEQ even in small doses acts injuriously on the human •animal organism. It is certainly true that a person may be poisoned by taking a sufiiciently large dose of burnt alum or of the crystallised solid alum, or even a large dose in concen- trated solution. It has also been satisfactorily established by Siemf that if animals are treated by subcutaneous doses of alumina salts a peculiar nervous condition may be produced, similar to the disease known as bulbar paralysis. On the other hand, it is a question whether in the moderate doses in which alum is taken in pastry or bread, or cakes, the flour of which has been mixed with alum or an alum baking powder, it has the slightest appreciable influence on health. The writer and the writer's family have used, ofl' and on, alum baking powders for years without injury. The fact that tons of alum baking powders are sold every week must show that a vast number of persons use alum baking powders, and yet no special malady has been recorded. To unintentional experiment there is added direct evidence, Christoph SchmitzJ gave a dog for three weeks aluminium acetate mixed with sausage; the dog took in the 21 days 99 grms. of alum acetate, and gained in weight half a kilogrm. To the same dog was given for 120 days, each day some 35 cc. of a solution of aluminium acetate, the total amount taken being equal to 260 grms. of aluminium acetate, and the dog increased in weight ,2 kilos. Schmitz himself took, for 31 days, 15 drops of the same solution, the total amount equalling 24 grms. of the solid acetate, and failed to find his health in any way aftected. Two young doctors took, for 33 days, 1 grra. of the tartrate of aluminium also without eff'ect. Schmitz carefully examined the Tii-ine of his experimental dog, but could not obtain more than feeble and doubtful evidences of a trace of aluminium hydrate, and he comes to the conclusion that aluminium compounds in moderate doses are not absorbed by the human intestine. I am, therefore, decidedly of opinion that alum in food in reasonable quantities is not injurious to health. * The use of alum is prohibited by the Bread Act, 6 & 7 Will. IV. c. 37. + Paul Siern, Ueber die WirJcung des Aluminiums und des Berylliums. Dorpat, 1806. X Christoph Schmitz, Untersuchungea iXber die etwaige Gijtigkeit des Aluminiums. Bonn, 1893. 204 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 90, On the other hand, the addition of alum to bread is inter- dicted by law, and whether alum is injurious or not does not touch the question of adulteration. Probably, in most cases, xmless the purchaser is expressly informed that alum is m such and such a cake or such and such a loaf, its presence should be considered an adulteration; for no one should be unwillingly compelled to take anything concerning which he has a prejudice against, whether that prejudice be founded on just grounds or otherwise. In other words, an alumed article of food will f'enerally fall under section 6, and not be of the nature and quality demanded. In searching for alum, the crust and the crumb should be analysed separately; for many bakers vise for the latter a flour technically called " cones," which is strongly alumed. and pre- pared from a fine species of wheat grown in the south of Europe, mixed with rice. This mixture is used for dusting the kneading trough and kneading boards ; in point of fact, for ^^ facing " the sponge previous to baking it. To search for alum in the crust, there is no other method save burning to an ash, as shortly to be described; but with regard to the crumb of bread, the qualitative test is the same as for flour — viz., an ammoniacal tincture of log- wood. From 300 to 400 grains of bread are crumbled in distilled water, and a slip of pure gelatin added, and the whole allowed to soak for twelve hours. On dissolving the gelatin in a little loc'wood, to which its own volume of a ten per cent, solution of ammonium carbonate has been added, if the bread is pure the solution will be reddish-pink ; if the bread is alumed, the solution will be blue, and exhibit the spectroscopic appearances described at p. 98. This blue colour is not absolutely decisive of alum, for bread adulterated with magnesia carbonate exhibits the same reaction; but if such a colour is produced, the bread requires further examination. , _ The author, in some special researches, has discovered that a certain portion of alum may always be washed out of bread as alum. He employs the following process : the bread is soaked in water for at least twenty-four hours (about -2 litres of water are used to 100 grms. of bread). The bread, is separated by means of a sieve, and the mass afterwards pressed in a cloth, ultimate clear filtration being obtained when necessary by aid of the mercury pump. This extract may be concentrated in a platinum dish, and when cooled a slip; of gelatin allowed to steep in a portion over night. The gelatin on being stained with logwood will exhibit a blue colour, if magnesia or alum is present. Another portion of the extract is dried and % 90.] BREAD. 205 burnt up in the usual way, as in the pi'ocess to be described, and the phosphate of alumina separated. The phosphate of alumina is now fused with sodic sulphate, the result of the fusion being sodic phosphate and alumina. The sodic phosphate is washed out with water; the alumina boiled with a drop or so of dilute sulphuric acid; to the sulphate of alumina thus obtained, a little solution of ammonia* is added, and the whole put in a watch- glass to crystallise over sulphuric acid. To obtain crystals in this way is often very difficult, but that alum is really present can be readily proved by the reactions of the solution with reagents. By strictly following these directions, a very small quantity of alvim can be detected. In a test experiment in a sample of bread in which 5 grains of alum had been added, it was found possible to obtain I'O in aqueous solution. The quantitative method for estimation of the total alumina in bread, as originally proposed by Dupre, and slightly modified by Wanklyn, is as follows : — 100 grms. of bread are incinerated in a platinum dish, until the ash does not exceed 2 grms. in weight. The ash is then moistened with 3 cc. of pure strong hydrochloric acid, and 20 to 30 cc. of distilled water added ; the whole is boiled, filtered, and the precipitate (consist- ing of unburnt carbon and silica) well washed, dried, burnt, and weighed. To the filtrate containing the phosphates, 5 cc. •of strong solution of ammonia are added. If the bread has been alumed, the phosphates now precipitated are those of lime, magnesia, iron, and alumina, of which the latter (viz., phosphate ■of iron and alumina) are insoluble in acetic acid, so that their separation is easy. The liquid is strongly acidified with acetic acid, boiled and filtered, and the phosphates of alumina and iron washed and weighed. Unless the liquid has been acidified sufficiently, phosphate of lime contaminates the precipitate and vitiates the results, so that this is an essential point. The last step is re-solution of the precipitate in acid, and the estimation of the iron ; this is usually best eftected by a colorimetric jirocess. A standard solution of metallic iron is made by dissolving a gramme of fine iron wire in nitro-hydrochloric acid, precipitating with ammonia, washing the peroxide of iron, and dissolving it in a little hydrochloric acid, and diluting accurately to 1 litre [1 cc. = 1 mgrm. of metallic iron]. On now adding to an tmknown very dilute solution of iron a known quantity of strong ammon. sulphide, a certain coloixr is produced, and this colour is exactly imitated in the usual way by a similar quantity of ammon. suli)hide and the standard solution, the whole opera- tion being conducted on the well-known principles of colorimetric * It is very easy to form a cesium alum, so that if, instead of ammonia, a, solution of ctesium chloride be added, the result is the formation of beautiful, well-marked crystals. 206 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 90. estimation. The amount of iron in the precipitate being known, it is calculated into phosphate, and the phosphate of iron sub- ti-acted from the total weight of the i)recipitate gives the •weight of the phosphate of alumina. From Mr. Wanklyn's experiments it would seem that in the case of bread ash, it is unnecessary to evaporate the hydrochloric solution to dryness, as is usually done, and that the separation of silica is complete by the method just detailed. Another perfectly valid way of estimating akimina in bread or flour, consists in a modification of the old Normandy process. The bread is burnt up as before, the ash powdered and treated •with hydric chloride, diluted with water, boiled, and filtered. The filtered solution is again boiled, and whilst boiling poured into a very strong solution of sodic hydrate, the whole boiled, filtered, and -washed. To the filtrate is added a few drops of disodic phosphate, it is then slightly acidified with hydric chloride, and subsequently rendered just alkaline by ammonia. The precipitate is collected, washed, and weighed as alumina phosphate. The following table will be of u.se in the conversion of phos- phate of alumina into alum : — ihate of Alumina, Ammonia Alum, Potash Alnm, Al203,P05. NH4,A1,2S04.1'^H20. KA1,2S04,1-.'H20. Parts. Parts. Parts. 1 3-7;^3 4-481 2 7-466 8-962 3 11-199 14-443 4 14-932 17-924 5 1S-6G5 22-405 6 22-398 26-886 7 26-131 31-367 8 29-864 35-848 9 33-597 40-329 10 37-336 44-813 If it is desired to separate the phosphoric acid, the phosphates of alumina and iron may be treated with six times their weight of sodic sulphate, as before stated. Since, when operating in the usual way, the alumina is not recovered as alum, but as a salt of alumina, it is of importance to know whether alumina is con- tained in unadulterated flour, and if so in what quantity. It is certain that properly cleansed wheat contains no trace of alumina; but pai'ticles of clay from the ground, as well as sand from the millstones, do as a fact get into wheat flour, and there is no second- class flour in commerce which does not contain some small per- centage of alumina. It might be expected that this adventitious alumina would have some sort of relationship to silica, for it may be presumed to exist as silicate of alumina. We fortunately § 90.] BREAD. 207 possess a few analyses by Dr. Dupre, and an elaborate research of Ml-. Carter Bell, which will very fairly settle the question. Dr. Dupre analysed twelve commercial samples of flour, none of which gave any reaction with the logAvood test, and the results of the quantities of alumina and silica are as follows : — Alumina, Silica, Per cent, of Ash. Per cent of Ash. Eatio. Minimiun, ■63 3-08 1 :4-7 Maximum, . 3-72 26-91 1 :7-2 Mean, . 1-98 10-4 1 : 5-2 Mr. Carter Bell analysed no less than forty samples of flour,. none of which contained alumina as alum, and the following are the main results : — Alumina. Silica. Katie. Minimum, •003 •009 1 :30 Maximum, •Oil •109 1:9-1 Mean, •004 •034 1 : 8-6 Mr. Carter Bell also analysed thirty-two samples of breads none of which gave any reaction with the logwood test ; the main results of these analyses are as follows : — Iron Alumina Water. Silica. Phospliate. Phosphate. Minimum, . 40-30 •010 -0005 •00^22 Maximum, . 49-50 -039 •0040 •0082 Mean, 45-56 -016 •0018 •0049 In these last researches with relation to bread, the ratio' between the silica and alumina is 1 of silica to 7^1 of alumina. If the alumina is translated into alum, the important result is obtained that the number of grains of alumina, if calculated into alum, about equals the silica. Thus, in the mean of the thirty-two samples of flour, the alumina was 4 mgrms. Now 4 mgrms. •004 of alumina is equal to •035 of ammonia alum, and the silica is •034. Or, again, if the mean numbers of the silica and alumina of the thirty -two samples of bread are taken, thei'e is •016 silica to -019 of the phosphate of alumina turned into akim. This, if calculated on the 4 lb. loaf, woiild be a little over 5 grains of alum. Hence from these researches it is clear, that in cases in which the analyst finds the presence of alum in bread from tests detailed, and then burns a quantity of the bread up, and in the ash estimates the phosphate of alumina and the silica, it will be a perfectly fair calculation, to allow for every part of silica found one part of alum, and this quantity is to be deducted as natural to the flour in the final calculatiori. 208 foods: tiikir composition and analysis. [§ 90. BIBLIOGRAPHY RELATIVE TO FLOUR AND BREAD. All tlie general treatises, without exception, enumerated at page 43 have articles or chapters on flour and bread, and may be consulted. Bala. — Farine de ble adulteree par de la farine de Lathyrus. Union Pharm., 1S61, t. ii., p. 151. _ Barrnel. — Sur une pretendue falsification du pain par les sulfates de cuivre et de zinc. Ann. d'Ni/g. et de Med Ug., 1832, t. vii., p. 198. Bell, J. Carter. — On Analysis of Flour and Bread. Analyst, vol. iv., 1879, p. 1-26. BiBRA, V. — Die Getreidearten u. das Brod. Nurnberg. 1871. Blyth, a. Wynter.— Note on "Alum in Bread," Analyst, June, 1878; Improved Processes for the Detection and Estimation of Alum in Bread and Flour, Analyst, Jan., 1882; Art. "Bread" in Dictionary of Hygiene. Lond. 187(3. BoLAs, Thomas.— C/tem. News, 1873; Dingler's Polytechn. Journal. Bd. 209, §. 399. Briois, C. a. — Sur le gluten et le melange de diffe'rentes farines. {These de 2)harmacie.) 1856. Chevallier. — Various papers on flour and bread adulteration. Ann. d'Byg. et de Med. Leg., 1840, t. xxiv., p. 82; ib., 1841, t. xxvj., p. 126 ; ih., 1842, t. xxvij., p. 306 ; ib., 1843, t. xxix., p. 39 ; ib., 1853, t. xli., p. 198; t. xliii., p. 171; t. xlix., p. 402; t. 1., p. 147. Journal d' Hygiene, May, 1878. Comaille.— ,/oim^ \ mh^ the form of " barley- ^^~ meal," the grain be- Cf I ing ground whole, ^ 'A and as pearl barley, *^^ yj^A the latter being the ;fjM iSBS^s*^" 'L'OiVi[iJ grain deprived of its ■j>J-'J^ coverings and round- "^ ed by attrition. Fig. 29 is a section of barley, x 190 : a is the Barley-meal in the layer of cells forming the outer coat of the seed ; time of Charles I. b, the inner ; c, the gluten cells, and d the starch- almost entirely took holding cells. B, barley starch, x 350. ^.j^g p^.-^^e of wheat as the food of the common people, especially in the north of England. The composition of barley-meal is as follows : — Per cent. Water, 15-06 Nitrogenous substances (albumen, I'O to 1 '7 per cent.), . ir75 Fat, 1-71 Carljo-hydrates (sugar, \-'2; dextrin, 1"7), . . • 70 '90 Woody fibre, '11 Ash, -47 The nitrogenous substances are, it would appear, similar to those of wheat, comprising albumen, gliadin, gluten-casein, gluten-fibrin, &c. The constituents of the ash of barley are as follows : — Per cent. Potash, 20-15 Soda, 2-53 Lime, 2-60 Magnesia, 8*62 Ferric oxide, ..... -97 Phosphoric acid, . . . . . 34-87 Sulphuric acid, 1"39 Silica. 27-64 Chlorine, ...... "93 Barley-meal is used as an adulterant of various foods, but in § 94.] RYE. 213 itself it is little tampered with. The detection of adulterations is mainly microscopical, and the dimensions and appearances of barley-starch are described at page 174. Barley Bread. — Barley bread, though biTt little used in Eng- land, is eaten in some parts of the Continent. The mean of two analyses by Von Bibra is as follows : — Per cent. Water, 12 -39 Nitrogenous substances, . . , 5 "91 Fat, -90 Sugar, 3-95 Carbo-hydrates, 71-03 Woody libre 5-63 RYE. § 94. The seed of the Secale cereale, in the form of rye-bread, was once a common article of diet in England, and it is now used as the daily bread of the northern European nations. The mean composition of rye flour is as follows : — Water, 14 Nitrogenous substances, ... 10 Fatty matters, 1 Sugar, 3 Gum, 7 Starch, 58 Woody fibre, 1 Ash, , 1 The nitrogenous substances in rye are made up of albumen, mucedin, and gluten-casein, but gluten-fibrin and gliadin do not appear to be present. The fat extracted from the rye has, according to Konig, the following composition : — Per cent. Glycerin, 1-30 Oleic acid, 91-60 Palmitic and stearic acids, . . . 8 '10 It, therefore, consists only in part of glycerides, some of the acids being in the free state.* The gum, according to M. H. Ritthausen, is soluble in alcohol, and has the ordinary composi- tion of gum. * The fat was saponified by lead oxide. (See the observations in the article on " Olive Oil.") !14 FOODS : THEIR COMPOSITION AND ANALYSIS, [§»^ The ash of the rye-flouv has, according to Y. Bibra, the follow- ing composition : — • Potash, Soda, . Lime, . Magnesia, . Ferric oxide, Phosphoric acid, Sulphuric acid. Chlorine, Per cent. 38-44 1-75 102 7-99 2-54 48-26 The composition of fresh rye-bread, according to twenty- seven analyses, from various sovirces, collected by Konig, is as follows : — Mean. Per cent. Per cent. Per cent. Water, 35-49 48-57 44-02 Nitrogenous matters, 3-49 9 22 6 02 Fat, ... 010 -S3 -48 Sugar, . 1-23 4-55 2-54 Carbo-hydrates, . 32-82 51-13 45-33 Fibre, . -29 •39 •30 Ash, . •86 3 08 1-31 None of the cereals are so liable to become ergotised as rye. [See author's work on " Poisons."] Roasted rye has been used to adulterate coffee, chicory, and other substances. It furnishes, by appropriate treatment, a good malt for the distillation of spirits, ■ and is used in the manufacture of Hollands. PJCE. § 95. Rice is obtained from the Oryza sativa, and the term is popularly ajjplied only to the seed denuded of husk and inner cuticle, the composition of which is as follows : — Per cent. Water, 14-41 Nitrogenous substance, 6-94 Fat -51 Starch, 77-61 Woody fibre, -OS Ash, -45 The oil which is obtained from the rice embryo has a density of ^924 at 15°, and at 5° becomes thick and buttery; it contains much olein and an albuminous substance.* * A. Pavesi, and E. Rotondi, Gazeita Chcmka ItaJiana, iv, 192-195. § 95.] RICE. 215 The composition of tlie ash of rice is as follows : — Per cent. Potash, 21-73 Soda, 5-50 Lime, 324 Magnesia, 11-20 Ferric oxide, 1-23 Phosphoric acid, 53-68 Sulphuric acid, '62 Silica 2-70 Chlorine '10 Rice is said to be adulterated from time to time with other starches, but it is in itself so cheap that it is more likely to be Fig. 30 represents the microscopical structure of rice. The figure is a section of the seed, x 190. a, is the outer hiisk , c, the gluten cells, and d, the starch- holding cells. B. The starch cells, x 350. used as an adulterant than tampered with. The microscopical examination of a sample would easily detect any foreign matters. The size and characters of the little granules have been already described at page 175, and are entirely different from all the other starches. A chemical method of detecting the falsification of rice has been proposed by M. Yan Bastelaer. It appears that a saturated solution of picric acid does not cause the least precipitate in a cold watery extract of rice; but if maize starch, leguminous starches, or other matters be present, there is a more or less abundant precipitate. The auantities recommended are 20 grms. of the powdered rice steeped for an hour in 100 grms. of water, and then the infusion decanted ; to this infusion the picric-acid test is applied. 216 FOODS : THEIR COMPOSITION AND ANALYSIS. [§96, MAIZE. § 96. Maize, or Indian corn {Zea Mays), a native of tropical America, is extensively cultivated in America, Africa, Southern Europe, Germany, and other countries. It is ordinarily met ■with as the India-corn meal of the shops, and forms the basis of many "infants' foods;" its use appears to be on the increase. According to analyses of A. Riche and A. H. Church, maize has. the following composition : — A. ElCHE. Per cent. Water, 17 'lO Albuminoids, 12-80 Starch, 59-001 Dextrin and Sugar, . . . . 1 '50 J Oil, 7-00 Cellulose, I'oO Ash, 1-10 According to Kbnig, the fatty matter of the maize contains 6-46 per cent, of glycerin, 79-87 per cent, of oleic acid, and 16-14 per cent, of stearic and palmitic acid. The mean composition of the ash from nine analyses of maize is as follows : — Per cent. Potash, 21-73 Soda 5 -51) Lime, 3-20 Magnesia, 11-20 Ferric oxide, 1-23 Phosphoric acid, 53-68 Sulphuric acid, ........ -62 .Silica, 2-74 Chlorine, -10 A. H. Chdrch.^ Per cent. 12-5 9-5 70-7 3-6 2-0 1-7 Fig. 31 represents a section of the seed, x 1 90 ; and B the starch, x 350. a is the outer husk ; h, the inner; c is the gluten cells ; d, the starch-holding cells. * The mean of several analyses of whole Indian grown maize. grains of India, London, 1886, 65. Food § 97.] MILLET. 217 An aqueous decoction of maize gives with a little iodine a peculiar reddish-purple colour. On placing the iodised decoction in the dark, after some eight to twelve hours the precij^itate becomes dirty white, and the supernatant fluid milky; if an excess of iodine is added the precipitate is then red, but it also becomes decolourised in the dark. Maize is said to be occasionally adulterated with potato starches, etc.; these a microscopical examination will at once detect. M. Genin has given certain chemical reactions, based upon the difterent hues which mixtures of maize and potato starch assume when treated with iodine, and also a process based on the volume which a precipitate obtained by lead acetate in an alkaline extract occupies in pure and adulterated samples. These processea are, however, far too loose to be of any service. MILLET. § 97. The millet seeds are derived from two species of panicum,, Panicum viiliaceum and Panicum italicum. It is extensively used among the Chinese and Eastern races as an article of diet, and its nutritive power is about equal to that of rice.* The average composition of millet deprived of its coverings, according to six analyses, is as follows (Konig) : — Per cent. Water, 11-26 Nitrogenous substance, 11-29 Fat, 3-56 Sugar, 1-18 Dextrin and gum, ...... 6"06 Starch, 60-09 Cellulose, 4-25 Ash, 2-31 The composition of the ash of the millet deprived of husk is as follows : — Per cent. Potash, 18-36 Soda, 3-82 Lime, Magnesia, ........ 21-44 Ferric oxide, 1-82 Phosphoric acid, 44-21 Sulphuric acid, 2-02 Sihca, 8-33 * In the author's "Dictionary of Hygiene" will be found a remarkable experiment on the nutritive qualities of millet. 218 FOODS : THEIE COMPOSITION AND ANALYSIS. [§ 98. POTATO. § 98. The chemical composition of the uncooked potato is, according to the analysis of some fair avei-age tubers, as follows : — Per cent. Water, 70-00 Starch, 19-68 Sugar, 1-20 Albumen,* '70 Gum, -40 Asparagin, '30 Fat, -30 Solanin, "05 Other nitrogenous substances, '15 Insoluble matter, '40 Ash, -82 inimum. llaximum. Mean. 68-29 'A'2 22 75-77 •51 3-60 1-79 •05 •80 •16 1205 26-57 20-56 •27 1-40 •75 •42 1-46 •97 100-00 A summary of seventy analyses, determining the principal constituents of the potato, is given by Konig as follows : — Water, Nitrogenous substances, Fatty matters, Starch, Woody fibre. Ash, . . It is thus seen that, according to all analyses, some 95 per cent, of the potato is water and starch. The nitrogen of the potato, which, according to the old method of analysis, would be reckoned into albumen, or, at all events, into protein substance, is derived from albumen, asparagin, solanin, xanthine, leucin, and tvrosin. It has been calculated that about 56 per cent, of the total nitrogen is derived from asparagin and amido acids — a fact •which must be remembered in diet calculations. Besides the constituents enumerated, there are certain organic acids in the potato which may be extracted in small quantities by sulphuric acid and ether. Among these are citric and succinic acids, and possibly the presence of these organic acids in part accounts for tlie antiscorbutic power possessed by the potato. Siewert has also found from -017 to -057 per cent, of oxalic acid. The composition of the ash of the potato, according to fifty- three analyses by E. Wolff, is as follows : — MiDimum. Potash, . . . . 43-95 Soda, Lime, '51 Magnesia, . . . . 1-32 Ferric oxide, . . . ^04 * The albumen, according to E. Schultze and E. Eupli^r {Landiv. Ver- euchs. ^iat. xxvii., 357), varies in dilierent kinds from -65 to ri9 per cent. Maximum. Mean. 73-61 60-37 16-93 2-62 6-23 2-57 13-58 4-69 7-18 1-lS § 98.] 219 Phosphoric acid, . Sulphuric acid, Silica, ..... Chlorine, .... Percentage of ash in dried substance. Minimum. Maximum. 8-39 •44 •85 2-20 27-14 14-89 8-11 10-75 5-30 17-33 6-49 213 3-11 3-77 The potato is very subject to a fungus disease, tho life-history of which has of late years been very fully elucidated by various observers, and more especially by Worthington O. Smith. The fungus is named botanically Pero- nos})ora infestaiis, and the manner in which it grows, and its method of reproduction, is shown in the annexed wood- cut (fig. 32). The figure re- presents the veiy highly magnified section of a potato leaf, and the mycelium of the peronospora growing among the cells. A, A are the natural hairs of the potato leaf; B, B are the upper and lower layers of the healthy cells. The threads and bodies at C, I), E, F, and G belong entirely ' x to the fundus J The fine thi'tiad ^- __^_- ^^ ^ _^ at C is a direct Fig. 3-. continuation of the spawn or mycelium living inside, and at the 220 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 98, expense of the leaf tissue. Emerging into the air, the thread is seen to bear two distinct species of fruit — one, D D, called simple spores or gonidia, while at E F are what are known as " swarm spores." The swarm spores, when moistened, set free fifteen or sixteen bodies, known as "zoo-spores," so named because they are endowed with spermatozoa-like motion, being furnished with two lash-like tails, which they move with great rapidity. A zoo-spore, when it falls on a leaf, has a surprising power of corroding the epidermis, and entering into the tissue. This action is probably due to some special solvent secreted by the zoo-spore. When movement ceases, the tails disappear, and a minute thread is protruded at one end, which develops into a network of mycelium. Both of tliese methods of production are asexual; but there is a third method which is sexual, and the more important, because, in this case, there are structures formed which resist frost and a variety of influences destructive of the former fragile structures. This third method is the production of oospores by the conjunction of two organs — the one named antheridium, and analogous to the anther of a flower, and the other oogonium, and analogous to the ovary of a flower. The oospores are dark -brown in colour, reticulated, and covered with little warty prominences about -001 inch in diameter.* Analysis of the Potato. — Since the potato is mainly composed of water and starch, a careful determination of those substances by the processes already enumerated will be sufficient for most purposes. For those in which great accuracy is not required, it will be sufiicient to take the specific gravity of the potato, and then refer to the following table. This table, for fairly good * The Peruvians make a national dish from frozen potatoes, which they call "chuno." The potatoes are steeped for a little while in water, then exposed for a few days to shar]) frost, washed, and rubbed. By this method the peel becomes detached, and the starchy matters are dried in the sun or in an oven. The dried hard tubers are cut in thin slices and baked, and eaten with the addition of Spanish pepper. An analysis by Meisel ( Wagner's. Jahresbericht der Chemisch. Technol., ISSl) is as follows : — Per cent. Water, 13 030 Starch, 81-844 mtroo-enous matter, . . 2-313 Woody fibre, . . . . MSB Fat, -182 Ash, -356 Soluble constituents in water, , 1'142 Total nitrogen, '4 %. Nitrogen soluble in water, -03 %. •400 sugar. •141 asparagin. •601 soluble starch, dextrin, soluble ash constituents, &c. 98.] 221 potatoes, will give results of from within '3 to '5 per cent, of the true value; but with regard to tubers poor in starch there may be a much larger error. TABLE XIII.— SHOWING THE PERCENTAGE OF STARCH AND DRY SUBSTANCE CORRESPONDING TO VARIOUS SPECIFIC GRAVITIES. Specific Dry Specific Dry gravity. substance. Starcli. gravity. substaDce. Starch. 1-080 19-7 13-9 1-120 28-3 22-5 1-081 19-9 14-1 1121 28-5 22-7 1-082 20 1 14-3 1-122 ■28-7 22-9 1-083 20-3 14-5 1-123 28-9 23-1 1-084 20-5 14-7 1-124 29-1 23-3 1-085 20-7 14-9 1-125 29-3 23-5 1086 20-9 151 1-126 29-5 23-7 1-087 21-2 15-4 1127 29-8 24 ros8 21-4 15-6 ri-28 30 24-2 1 089 21-6 15-8 1-129 30-2 24-4 1090 21-8 16-0 1-130 30-4 24-6 1-091 22-0 16-2 1-131 30-6 24-8 1092 22-2 16-4 1-132 30-8 25-0 1 093 22-4 16-6 1-133 31-0 25-2 1-094 22-7 16-9 1-134 31-3 25-5 1-095 22-9 17-1 1-135 31-5 25-7 1-096 23-1 17-3 1-136 31-7 25-9 1-097 23-3 17-5 1-137 31-9 26-1 1-098 23-5 17-7 1-138 32-1 26-3 1-099 23-7 17-9 1-139 32-3 26-5 1-100 24-0 18-2 1-140 32-5 26-7 1-101 242 18-4 1-141 32-7 27-0 1-102 24-4 18-6 1-142 33 27-2 1-103 24-6 18-8 1-143 33-2 27-4 1-104 24-8 19 1-144 33-4 27-6 1-105 25 19-2 1-14$ 33-6 27-8 1-106 25-2 19-4 1-146 33-8 28-0 1-107 25-5 19-7 1147 34-1 28-3 1-108 25-7 19-9 1-148 34-3 28-5 1-109 25-9 20-1 1-149 34-5 28-7 1110 26-1 20-3 1-150 34-7 28-9 1111 26-3 20-5 1-151 34 9 29 1 1-112 26-5 20-7 1152 35-1 29-3 1-113 26-7 20-9 1-153 35-4 29-6 11 14 26-9 21-1 1-154 35-6 29-8 1115 27-2 21-4 1-155 35-8 30-0 1-116 27-4 21-6 1-156 36 30-2 1-117 27-6 21-8 1-157 36-2 30-4 1-118 27-8 22 1-158 36-4 30-6 1-119 38 22-2 1-159 36-6 30-8 222 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 99. PEAS. § 99. The pea is, without doubt, the most important of all the leguminous plants. The gai-den ]iea is derived from the Fisum sativum, a native of the south of Europe, but long naturalised in this country. The field pea, grown for the purpose of feeding cattle, is the Pisum arvense. Forty-one analyses collected, by Ivonig give the following values : — ■ Maximum. Minimum. Mean. Water, 22-12 11-01 14-31 *Nitros!enous substance, 27 14 18-56 22 03 Fat, 3-30 •64 1-72 Nitrogenous free extractive matter. 59-38 41-90 53-24 Woody fibre, .... 10-00 2-22 5-45 Ash 3-49 1-76 2-65 The 53-24 per cent, of non-nitrogenous soluble matter is com- posed of 36-03 starch, 5-51 dextrin, and 11-70 other substances, among which is some sugar. Cholesterin is also found in peas, but there have been no researches as to its exact quantity. The most important principle of the pea is " legumin." Its amount varies in different species. Thus, H. Ritthausen found in the green field pea, 3-95 per cent.; in the yellow, 9-45, and. in the grey, 7-30 per cent.; in tbe garden pea, 5-40 per cent. In the young unripe condition, peas contain much more water than the proportions given above. Thus, Grouven found, in young unripe peas and beans the following : — Grepn Peas. Green Beans. Water, 79-74 9134 Carbo-hydrates, .... 13-03 5-99 Albuminoids, .... 606 2-04 Salts, 1-12 -63 The albumen of peas (which may be obtained by boiling the solution after it has been freed from legumin) differs from ordinaiy vegetable albumen, both in elementary composition and in its behaviour to reagents. An analysis gives the following percentages: — C, 52-94; H, 7-14; N, 17-14; S, 1-04. After coagulation it dissolves in potash water to a clear liquid, which is not the case with vegetable albumen. The analysis of the ash of peas gives the following as the extremes and mean of twenty-nine analyses : — * Some nitrogen determinations by C. Bohner (Versuch. Stat, xxviii., 247-2(52), in which ammonia, amido aeids, and other products were ditler- entiated, are as follows: — Total nitrogen in dried peas, 4-69 per cent.; nitrogen as albumen, 3-56 (=22-5 per cent, allniminoids); nitrogen as ammonia, -020; nitrogen as amido acid amides, -052 per cent.; nitrogen as amido acid, -361 ; nitrogen as peptones, -697 per cent. 99.] 223 Potash, Soda, . Lime, . Magnesia, . Ferric oxide, Phosphoric acid, Sulphuric acid, Silica, . Chlorine, Minimum. Maximum. Mean. 35 80 51-41 42-79 3-57 -96 2 •21 7-90 4-99 5-80 13-U2 7-96 3 83 •86 29-30 44-41 36-43 9-46 3-61 3-02 •86 ... 6-50 1-54 Peas, when putrid, undergo some peculiar change not yefe investigated, resulting in the formation of a poison, perhaps similar to the cadaveric poisons described in the author's work on " Poisons."* For the general analysis of peas, the water, the ash, and the amount of starchy matters are estimated by the processes already detailed. To separate the legumin, the peas must be powdered, or, if fresh, mashed into a paste, and treated with successive quantities of cold water, which may be advantageously feebly alkaline, but must not have the least trace of acid. The legumin may now be precipitated by acetic acid, the precipitate dissolved iu weak potash, again precipitated, and then dried and weighed. * Not very long ago a case of wholesale poisoning from this cause occurred in Salford. Many persons who had partaken of slightly decomposed peas exhibited symptoms of irritant poisoning. The peas were chemically examined, but contained neither arsenic, copper, lead, or other metallic poison. [Pliar. Joiirn. (3), 294.] The subject of the formation of new and poisonous substances in such an article of food would well repay investiga- tion. In Germany there has been used a condensed food made up of powdered and dried meat, incorporated with pea-meal, by strong pressure ; it is scarcely necessary to say, that in this manner a food invaluable for the soldier is obtained, and oue that contains in a very small compass all the essentials of nourishment. An analysis of these pea-meat tablets is as follows : — Per cent. Water, 1209 Nitrogenous matters, 31-18 Fat, 308 Carbo-hydrates, 47 '50 Ash, 615 A condensed pea soup is also prepared. Two analyses of this condensed soup, given by Konig, are as follows : — Water, Nitrogenous matters. Fat, . Carbo-hydrates, . Woody tibre. Ash, . 1. 7 -58 16-93 8-98 53 44 1-34 11-73 2. 8-08 15-81 24-41 36-78 1-69 13 23 224 FOODS : their composition and analysis. [§ 100. Leguuiin is almost insoluble in cold or warm water ; hut since it may be extracted so easily from the fresh seeds, it is supposed to be in combination with phosphates of the alkalies when in its natiii-al condition. But it is easily soluble in diluted alkaline liquids, and also readily dissolves in a solution of alkaline phos- phates; if boiled it becomes insoluble in alkalies. Pure alkaline solution of legumin shows, with a little cupric sulphate, a beauti- ful violet colour. If impurities are present, such as gum or starch, the colour is blue. On boiling the alkaline solution, the legumin. does not coagulate, but, as in the boiling of milk, a scum of altered legumin appears on the surface. § 100. Preserved Peas. — Copper iit Peas. — Peas are preserved in ■several ways, sometimes by simply drying, when they form the well- known dried peas of the shops. But the more modern method is to heat the peas in a suitable tin capable of being hermetically sealed. The sealing is effected while the tin with its contents is at a high temperature. The rationale of the process is, that putre- fying germs existing on the surface of the peas are destroyed, and fresh putrefactive agencies are prevented from gaining access by the exclusion of air. Peas so pi'eserved may, as proved by ana- lysis, be quite as nutritious as fresh peas. Preserved peas have often undergone a preparatory treatment by boiling in copper vessels, the object of which is to impart a fine green colour. The reason why vegetables preserve their green colour when treated by cop])er is, according to Tschirch,* owing to the formation of a compound of copper with phyllocyanic acid (C04H28N2O4). Phyllocyanate of copper has the composition (C24H27No04)oCu, and contains S'55 per cent, copper; it is insoluble in water, but soluble in strong alcohol and chloroform ; it is not soluble in dilute acetic nor in dilute or concentrated hydrochloric acid. Its alcoholic solution gives an absorption spectrum characterised by four bands. Tschirch has shown that although copper phyllocyanate is so insoluble in acids, yet when administered to animals it produces the same eflfects as equal quantities of copper given as the soluble tartrate ; at the same time' he considers that so small a quantity of copper added as will produce phyllocyanate only would probably not be injurious to health. M. Guillemare and M. Lecourt have, however, now patented a process by which chlorophyll has been substituted for the objectionable coppering. The copper that has hitherto been found in tinned peas has amounted to about *2 grains to 2-6 grains in the pound tin, and the question arises whether the copper is injurious to health in this ]iroportion or not. In the cases appended to this article it will be noticed that men of considerable scientific reputation have expressed strong opinions on the subject ; never- theless, the whole of the injurious action of coppered peas rests entirely on theory, and in no single instance (although the consumption of coppered peas has been very large) has any really definite case been brought forward of actual poisoning by peas coloured in this way. Legrif has found in the intestine of a healthy man •036 to -040 grm. of copper ;t and Messrs. * Das Kupftr. Stuttgart, 1S93. t Sonnenschein. § 100.] PEAS. 225 Paul and Kingzett have shown that, even when a soluble compound, like sulphate of copper, is ingested, most of it is excreted by the fsces. Experiments are not wanting which might be interpreted that copper was in no way harmful ; for instance, Dii Moulin {J. F/iarm. [5] 13, 189, 190) has dosed dogs witli from 3 to 5 grms. of copper sulphate, and induced vomiting without after-deraugement of health. He gave doses of from i to ] grm. of copper subacetate every day for six weeks to dogs and rabbits ■without producing injury. Carbonate and oxide of copper he also adminis- tered to rabbits without hindering growth. Compoimds of copper with fatty acids gave similar negative results. On the other hand, it must be remembered that small doses of copper have produced serious symptoms. The whole question, indeed, of botii copper and lead poisoning is very obscure, and the existing evidence abounds with paradoxes. The author, therefore, concludes that it would be best, in our present state of knowledge, to decline to state whether cojjper, when it exists in the moderate proportions given above,* would be likely to injure health, since there are no definite facts iipon which to base a sufficient opinion. The experiments of Tschirch would seem to show that it is important in researches on copper to divide the copper found into that obtained from (a) an alcoholic extract and (b) insoluble in alcohol. In any case coppered peas will not be of the nature and quality demanded by the purchaser, and, therefore, may be certified as adulterated under the Sale of Food and Drugs Act. The method of detecting copper in peas is as follows : — A weighed quantity of the ])eas is made into a paste with water and a little hydric chloride, and the paste is placed in a ])roper platinum dish ; a rod of zinc, on being inserted in the paste, .so as to touch the platinum disli, sets up a galvanic current, and at the end of several hours all the copj)er is deposited as a coherent tilm, and may be dried and weighed. A neater process is the connection of the platinum dish with the negative pole of a battery, while the positive pole is suspended in the acid paste. In both instances the copper is deposited as copper. Tschirch finds that rnicoppered chlorophyll extracted with alcohol, the alcoholic solution evaporated, the extract washed with water and treated with concentrated HCl gives a deep blue phyllocyanin solution, and a residue dissolving in ether with a brown-yellow colour ; coppered chlorophyll, on the contrary, gives a yellow, not blue, solution, and the residue dissolves in alcohol with a green colour. If instead of concentrated HCl, dilute HCl is added to alcoholic extract, uncoppered chlorophyll dis- solves with a yellow, coppered chlorophyll with a green, colour. Tinned peas may contain traces of tin. The process for the detection of tin is as follows : — A sufficient quantity of the peas is incinerated in a ■|)latinum dish, the ash is heated with strong hydric chloride, and evaporated nearly to dryness ; a little water is then added, boiled, and the solution filtered. This method of extraction is repeated once or twice. The solu- tion is now saturated with hydric sulphide, and any yellow precipitate filtered off. This should present the characters of sulphide of tin. Tin has been found, according to Mr. Hehner,t generally in tinned goods to the amount of 10 mgrms. in the English pound, and it has been supposed, without adequate proof, to exist as a stannous hydrate, a tin compound which is poisonous. * Analyst, 1877, p. 98. iAnal>/tit, 1880, p. 218. 16 226 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 100. No prosecution has hitherto taken place with regard to tin in. preserved goods: and in such small quantities as have hitherto been found, it is very questionable — presuming the tin to exist as stannous hydrate — whether any injury would result. A few of the more important prosecutions for coppered peas may, in conclusion, be quoted: — In one of several similar cases heard at the Marlborough Street Police Court, it was proved by Mr. Piesse, the analyst, that the pound tin contained ■088 of copper, equal to 2^ grains of sulphate of copper. Dr. Conway Evans, Medical Officer of Health, stated his opinion, that the larger quantity of copper spoken of in a pound of peas, if eaten daily or re- peatedly, wouki be injurious to health, and would produce chronic poisoning; but many persons might eat a quantity of these peas several times without apparently suffering any injurious effects, the period varying in accordance with difi'erence of vigour, age, health, &c. Two or three doses might affect some ]iersous and not others. From 14: to 15 grains of copper were some- times given as an emetic ; and sometimes, in ague or chronic diarrhoea, ^ ta 3 grains were given as a tonic. It was a well-known medical fact, that in regard to some poisons (such, for instance, as mercury) certain persons were peculiarly susceptible to their influence ; and it was possible that these peaa containing copper, if swallowed by persons ignorant of their own suscepti- bilities, might (even in a single dose, or a few doses) lead to injurious con.se- quences. He believed copper to be more fatal, in a smaller dose, than salts of lead. The heightening of the colours of preserves with copper was once a common practice. Cases of poisoning by copper were formerly very common, but copper utensils in cooking having given place to tin and iron saucepans, such cases were of rare occurrence. Pure metallic copper he believed to be harmless, but it was dangerous when in contact with other substances, and when dissolved. Dr. Guy, F.Pi.S., said that cases of poisoning by copper had occurred in which the quantity swallowed must have been small. He had studied the question of poisons particularly ; the fact of a trace of copper in the human body would not i)rove its existence in a poisonous form. He had made inquiries for Government into the effects of poisoning in certain trades ; palsy followed the poisoning by copper. Two cases had come under his knowledge of poisoning by the green wall-paper in a room ; the poisoning in, his opinion came from the copper, not the arsenic. Salts of copper he con- sidered more poisonous than lead ; the small quantity of copper contained in the peas in question from France might prove injurious, and slowly undermine health. On a nervous person copper was more likely to [jroduce dangerous symptoms than on any one else. ^Vith regard to the presence of 3-6 of copper, if taken one-third at a time, it would not affect a healthy person ; but if repeated in small doses it would, in his opinion, be ultimately injurious to health. He considered that any article containing the amount of copper stated by Mr. Piesse should not be allowed to be sold for one moment. Sulphate of copper in its virulence ranked fourth in the class of poisons. Dr. C. Tidy gave similar evidence. If copper, that is, sulphate of copper, were constantly taken to the extent of the amount of copper found in the French peas, it would be injurious to health. Dr. A. Dupre stated that the quantity found was far beyond the quantity normally present in any vegetable. Dr. Gu3' said he considered the sale of an article containing such a quantity § 101.] CHINESE PEAS. 227 of copper as that found in the French peas ought not to be tolerated. Small doses of copper were more dangerous than large ones, as the latter would cause vomiting. A previous conviction against the defendant for the same offence was proved, but the prosecution stated that they desired publicity, not punish- ment, and a small fine was inflicted.* At the Liverjiool Police Court a firm was prosecuted for selling peas containing copper equal to 24 grains of sulphate to the pound tin. A warranty produced, t A Liverpool grocer was fined 20s. and costs for selling peas containing copper equal to 2 6 grains of the sulphate to the pound. J At Bradford vendors have been fined for selling coppered peas, the metal equalling 1^ to 2 grains to the pound. § CHINESE PEAS. § 101. A pea or bean, mnch used in China in the form of cheese, is the Soia hispida.\\ Its composition, according to G. H. Pellet, is as follows : — 1. 2. X Water, 9-000 10-160 9-740 Nitrogenous matters (coagulable nitrogen, 6-25), . . . 35-500 27-750 31750 Starch, dextrua, and sugar, . 3 210 3-210 3-210 Cellulose, ITGoO 11-650 11-650 Ammonia, -290 -274 -304 Sulphuric acid, .... -065 -234 -111 Phosphoric acid, .... 1-415 1-554 1-631 Chlorine, "036 '035 '037 Potash, 2-1.37 2-204 2-317 Lime, -432 -316 -230 Magnesia, '397 "315 -435 Substances insoluble in acids, . -052 -055 -061 Not estimated mineral substances, -077 '104 -247 Different organic matters, . . 15-289 25-539 24-127 * Analyst, 1877. + Sanitary Record, vi. 335. X Sanitary Record, vi. 351. § Ibid., vii. 6.'^. II The pea-cheese is considered, in China and Japan, a very important food. The peas [S(na Idspida) are soaked in water for about 24 hours, then strained ; they are next ground to a thin jJaste with some of the water which has been put on one side. The grinding is effected by a mill. The matters are filtered, and the filtrate is concentrated by heat ; and after skimming once or twice is cooled, the casein coagulated by plaster, and a salt, which appears to be chloride of magnesium, added. The cheese is grayish-white, and has the following general composition : — Per cent. Water, 9037 Fatty matters, 2 36 Nitrogen, -78 Ash, -76 — M. Stanislaus Julien et M. Paul Champion, '■'Industries de r Empire Chinois." 228 FOODS : their composition and analysis. [§§ 102, 103. LENTILS. § 102. The lentil is the seed of the Ervum lens, one of the Legiiminosse. Lentils are grown and eaten in all parts of the civilised world, and are highly nutritious. They contain, according to H. Ritthausen, 5-9 per cent, of leguinin, and their general composition is as follows : — Per cent. Water, 12-51 Nitrot;enous substances, 24-81 Fat, \ 1-85 (Jarbo-liydrates, 54*78 Woody tibre, 3-58 Ash, 2-47 The genei'al composition of the ash is as follows : — Per cent. Potash, 34-76 Soda, 13-50 Lime, 6-34 Magnesia, 2*47 Ferric oxide, 2-00 Phosphoric acid, 36-30 Chlorine, 4-63 BEANS. § 103. The beans eaten in this country are mostly the kidney bean, Phaseolus vulgare, and the broad bean, Vicia faha. The following is the average composition of these vegetables : — Broad bean. Kidney bean. Water, 14-34 13 60 Nitrogenous substances, . . . 23-66 23-12 Fat 1-63 2-28 Carbo-hydrates 49-25 53-63 Woody tibre, 7*47 3-84 Ash, 3-15 3-53 The percentage composition of the ash of these difterent beans has the following composition : — Bro'id bean. Kidnev bean. Potash, 42-49 44-01 Soda 1-.34 1-40 Lime, 4-73 6-38 Magnesia, 7-08 7-41 Ferric oxide, -57 '32 Phosphoric acid, . . . . 38-74 35-00 Sulphuric acid, .... 2-53 4-05 Silica, -73 -57 Chlorine, 1-57 '86 § 103.] BEANS. 229 From both the broad and the kidney bean a small qxiantity of cholesteiin can be separated. According to Ritthausen, the legumin of the kidney bean has a composition difterent from that of other legumins ; for while the percentage of nitrogen in 2>ea and millet legumin amounts to 16-77 per cent., that of kidney beau legumin has only 14-71 per cent. PART IV.-MILK, CREAM, BUTTER, CHEESE. MILK. HISTORICAL INTRODUCTION. § 104. Before the birth of experimental philosophy, the origin rather than the composition of substances was the subject of inquiry, and of fanciful and moi'e or less ingenious conjecture. Milk to the ancient, as well as to the modern world, was a fluid of great virtue. Aristotle affirmed, "Zac est sanguis concoctus, non corruptus," which may be translated. Milk is elaborated, not decomposed, blood — an opinion identical with that held by nine- teenth-century philosophers. Averroes, Avicenna, and others, reasoning in part from the difficulty with which many females conceive while suckling, held that milk was altered menstrual blood. Avicenna, indeed, for- nuilarised this doctrine by declaring that the menstrual blood of the pregnant was divided into three parts — part going to nourish the fcetus, part ascending to the breasts, and the remainder be- ing an excrementitious product. These opinions may be traced to writers of a much later, almost modern epoch. The ancients were acquainted with only three constituents of milk — viz., buttei', with which they used to anoint their infants ; casein, which they precipitated with vinegar; and the whey from which the curd and butter had separated, and this, up to the early part of the sixteenth century, constituted the whole of what was known as the composition of milk. Placitus enumerates no more constituents than Avicenna, but devotes several pages to the then all-important question as to whether milk was hot, cold, or moist, and concludes that animal milk, as compared with that of human, is cold, human with that of animal, hot. Placitus* was an up- holder of the menstrual theory. Panthaleonj similarly cites with approval the dictum that milk is a fluid superfluity, tioice concocted in the breasts, and gravely discourses, as stated, whether it is hot * Sexti Placiti Papyriensis : De Natura et Um Lactis, mdxxxviii. It would appear, according to this author, that the Germans in his time used the milk of all animals, for he enumerates the milk not only of cows, mares, and goats, but also of pigs. t bumma Laciicinorum, 1528. § 105.] MILK — HISTORICAL INTRODUCTION, 231 or cold.* He recognises three parts only in milk — viz., serum, battel', and curd. His treatise is mainly composed of references to the ancients, and the usual disputations as to whether milk is hot or cold. The first mention of a fourth constituent of milk occurs in a curious work b}'^ Barfcoletus, published in 1619. Bartoletusf called it the "manna" of milk, or "nitrum serilactis." In his days svilphur, mercury, and a saline principle, were con- sidered as the three active essences of all things, and as exist- ing in all things ; hence, Bartoletus, from the yellow colour of butter referred it to a suli)hur principle, the whey, doubtless from its mobility, to quicksilver, and the curd to a saline element. He then compares milk with blood, also composed of a sulphurated, saline, and mercurial principle. | The discovery of Bartoletus for a long time was not known beyond Italy. A French apothecary, named Bartholomew Martin, writing in 1706,§ enumerates the con- stituents of milk as three — butter, analogous to sulphiir, serum to mercury, and cheese to salt ; but was not acquainted with milk- sugar, although eight years before Ludovico Testi|| had written an entire treatise on it, calling it by the name it now bears. In the early part of the eighteenth century, Leeuwenhoek discovered the microscopical characters of milk. He saw that it was a fluid containing many globules. Some, which he judged to be of a buttery nature, rose to the top of the liquid; and others, again, rather sank to the bottom, and were evidently diflerent in composition.^ Some twenty years later, A. Donne, in his Cours Microscopiqne** jtublished some beautiful plates of several kinds of milk, fresh and sour, human and animal, exhibit- ing the globules, &c., drawn to scale with wonderful accuracy. § 105. In the early part of the eighteenth century flourished the school of the illustrioiis master Boerhaave, who laid the * There are several other treatises on milk about this epoch, but they nearly all, as, for example, that of Gesner (Libellus de Locte et Operibus Lactarius, auth. Conrado Gesnero, Medico), consist of commentaries on the opinions of older writers, aud are of no value. t Bartoletus was an Italian physician, a professor at Bologna and Mantua, B. 1586, D. 1630. His work is entitled, Encyclopcedia Hermetico-Dorimatica sive Orbis Doctrinarian, Physiol,oraktiscliK Chemie, vol. 6, p. 1, 1S7'2. X Hoppe-Seyler, Physiologische Chemie, p. 930. § 115.] THE COMPOSITION OF COWs' MILK. 243 rennet, is th'e splitting up of the casein into two bodies, one of which is precipitated, and an albuminoid, wliich remains in solution, and is neither precipitated by boiling, nor by any of the following reagents — -acetic acid, potassic ferrocyanide, or nitric acid ; but is precipitated by mercuric chloride, and also by Millon's reagent. Casein is precipitated by a variety of substances — lead acetate, cupric sulphate, alum, mercuric chloride, tannic acid, rennet, suli)hate of magnesia, and mineral acids, if not too dilute ; but none of these precipitate casein in a pure state, the precipi- tate usually containing fat, nuclein, and phosphate of lime, the lattei", as already stated (p. 23G), in the proportion of from 1 to I'D per cent, of casein. The best precipitant is sulphate of magnesia, which leaves the nuclein to a great extent in solution. The fat may then be extracted by ether ; but the phosphate of lime is in true combination with the casein, and only a portion of it can be removed. A solution of casein in combination with sulphate of magnesia, and freed from fat, turns a ray of polarised light in weak alkaline solution, — 87° ; in very dilute alkaline solution, — 87° ; in strong alkaline solution, — 91°. Pure casein is a per- fectly white, brittle, transparent substance, insoluble in water, but soluble in very dilute acid solution, as well as in very dilute alkaline solution ; in each case there is little doxibt that a true chemical combination is formed. The presence of phosphate of soda in a solution of casein (as, for example, in the milk itself), prevents the px'ecipitation by simple neutralisation by an acid, the casein not falling down until the acidity of the liquid is decided. It has been shown by Schutzenberger that, on sealing up casein in a tube and heating with bar3'ta water, it behaves like albumen, and is resolved into the following substances : — The elements of urea (ammonia and carbon dioxide), traces of sulphurous acid, of sulphuretted hydrogen, of oxalic and acetic acids, tyrosin, CgH^jNOg, the amido-acids of the sei'ies C^Hoq^jNOv,, corresponding to the fatty acids, CaHq^O,, from amido-cenanthylic acid to amido-proi>ionic acid — leucin, CgH-,^3N02, butalanin, C^Hj^^NOg, and amido- butyric acid, C^HyNOg, with a few less known or identified products. Many of these substances may be identified in putrid milk. The amount of casein in milk is fairly constant, being about 3-9 per cent. ; and the author has never known it exceed 5 per cent. Serum- albumen occurs in milk, in no respect difi'ering from the albumen of the blood. By careful addition of an alkali, this albumen may be changed into alkali-albuminate — that is, into casein ; therefore, according to this view, the albumen in milk 244 foods: their composition and analysis. [§115. may be considered the residue of an incomplete reaction. Albu- men is not precipitated by acetic, carbonic, phosphoric, or tartaric acids. A small quantity of a dilute mineral acid does not pre- cipitate ; with a larger quantity of concentrated mineral acid the solution becomes troubled, and the deviation of a ray of polai'ised light increased ; a still larger quantity of acid precipi- tates it as acid albumen. The best method to obtain a solution of pure albumen is to precipitate a solution by basic acetate of lead, pass carbon dioxide through the mixture, separate the carbonate of lead by filti-ation, and, lastly, pass through it hydric sulphide, to remove the trace of lead still existing. Albumen is then in solution, but with a little acetic acid, on evaporation, it may be obtained in the solid state contaminated slightly with acetic acid."'' Another method of obtaining albumen pure is by dialysis. The physical characters of solid albumen differ according to the method oi separation. Albumen obtained by dialysis is in the form of a •^•ellow transparent mass, specific gravity 1"314; but albumen separated in the ordinary way from milk, for the purpose of quanti- tative determination, is in yellowish flakes, brittle, without taste or smell, insoluble in water, alcohol, and ether, soluble in dilute caustic alkali, if gently warmed, and from this alkaline solution precipitable by an acid. The amount of albumen in milk is really faiidy constant, and averages '7 per cent. In healthy cows it is a very constant quantity, the chief deviation occurring dii'ectly after calving, when the amount may rise as high as 3 per cent., but this is always accompanied by a corresponding rise in the casein. According to tiie author's experience, t])e albumen preserves a very constant relation to the casein, the quantity of the latter being five times that of the albumen; so that if either the amount of casein or albumen is known, the one may be calculated from the other with great accuracy. Kuclein. — Nuclein is the organic phosphorus compound of milk, containing, according to Miescher, 9'G per cent, of phos- phorus. Its formula is CggH^gN^P^Ooo. It is by no means peculiar to milk, but has been found in the blood, in pus, in the yelk of eggs, in the liver cells, and in yeast cells. When freshly precipitated, it is a white amorphous body, somewhat soluble in water ; freely soluble in ammonia, soda solution, and phosphate of soda. The special test distinguishing nuclein from other albuminoids is the presence of phosphorus, and the production * Meggenhofen appeai-s to have been one of the first who fletected the presence of albumen in milk. He estimated the amount in cows' milk as 'SO per cent. Dissertatio Inauguralis sistens indarjationem Lactis Mulitbris Chemkam. C. Aug. Meggenhofen. Frankfort, 1826. § IIG.] THE COMPOSITION OF COWs' MILK. 24 5 of no red coloui-, either by Millon's reagent, or by a copper salt, added to a solution of nuclein alkalised by soda lye : it forms a very definite compound with lead, the lead and phosphorus being in the proportion of Pb to P. The method adopted by Hoppe-Seyler* to separate niicleiu from pus, was isolation of the pus cells by Glauber's salts, washing with very dilute hydro- chloric acid and much water ; then extracting the nuclein by the aid of a very weak alkaline solution of caustic soda, and tiltei'ing (which in this case proves a troublesome operation), and precipitating by a mineral acid. The precipitate is again dissolved in weak alkaline solution, and again precipi- tated, and the process repeated until the nuclein is supi)osed to be in a fairly pure condition. Nuclein may be separated from milk on the same princii^les, tirst exhausting the solids by alcohol and ether to remove fat. § 116. mik-Sugar, CioHooOnH.p.— Milk-sugar, so far as is known, is only found in human milk, the milk of the herbivora, and of the bitch. It is easily distinguished from other sugars ; its specific gravity is 1-53 ; and its solution turns a ray of polai'- ised light to the right at 20^ 53°. The separate researches of Erdmannf and SchmcegerJ have shown the existence of four modifications of milk-sugar, exhibiting a different rotation to normal sugar, viz. : — 1. Crj'stallised milk-sugar exhibiting in solution .strong bi-rotation. 2. Anhydrous milk-sugar obtained by dehydrating crj'stallised sugar at 1.30°, showing in solution also strong bi-rotation. 'S. Anhydrous milk-sugar obtained by quickly evaporating a solution of milk-sugar in the presence of sand, or other finely divided substance, so as to ensure a large surface; the solution shows slight bi-rotation. 4. Anhy- drous milk-sugar evaporated down from a solution quickly, but without the addition of sand or other substance. All these modifications are at once transformed into milk-sugar of normal rotation by boiling their solutions, or gradually, without the application of heat. Milk-sugar is soluble in six pai'ts of cold, and 2"5 parts of boil- ing water ; it is insoluble in absolute alcohol and in perfectly dry ether, but in dilute alcohol and commercial ether it is slightly soluble, the solubility in amount depending mainly on the per- centage of water which the ether contains. At 150° it loses an atom of water without fui'ther decomposition ; its watery solution is perfectly neutral, and has a sweet taste : the sweetening power of milk as compared with cane sugar is but feeble. It reduces Fehling's copper solution in a proportion different from that of grape-sugar (see p. 139). Milk-sugar undergoes lactic fermentation readily (see p. 251), but alcoholic with some difficulty. Milk-sugar is precipitated by acetate of lead and ammonia; neutral acetate of lead, even at a boiling temperature, neither precipitates nor changes it. The oxides of copjjcr, of bismuth, and silver are reduced by soluticms of milk-sugar, and * Med. Cliem. Untersvch., Hoppe-Eeyler. Berlin, 4 Heft. t Beric/it, xiii. 2180-21S4. Z Ibid. xiii. 1915^1931. 246 foods: their composition and analysis. [J 117, indigo is decolourised ; these latter reactions are similar to those of grape sugar. When oxidised by nitric acid, milk-sugar yields mucic acid, acetic acid, and tartaric acids, and on further decom- position oxalic acid may be obtained. By boiling milk-sugar for several hours with 4 parts of water and 2 per cent, sulphuric acid, neutralising with carbonate of lime, evaporating the filtrate to a syrup, a different sugar from lactose may be obtained in microscopical crystals. To this altered milk-sugar, the name of galactose has been given. Its action on polarised light is expressed as -f 83-22 at \o° ; it is a fermentable sugar, and yields, on oxidation with nitric acid, twice as much mucic acid as milk-sugar. The amount of milk-sugar in normal milk preserves a very constant relation to the percentage of c;isein, being about -1 grm. per every 100 cc. in excess of the casein. Its average is about 4 per cent. § 117. Mineral Constituents of Milk. — The mineral constituents of milk have been fully and early investigated, and the following may be considered a very close approximation to their actual amount and character : — Potassium oxide, K„0, 1S*S2 Sodium oxide, NaoO, . . • . . . 11 "58 Calcium oxide, CaO, . . . . . . . 22 '97 Ferric oxide, FejOj "06 Chlorine, CI, 16-23 Magnesium oxide, MgO, 3-31 Phosphoric pentoxide, PjOj, 27-03 Four analyses of milk ash by R. Weber and Haidlen give the following : — Minimum. Maximum. Mean. 25 24-67 18 9-70 55 22-0 10 3 05 76 -53 13 28-45 •30 14-28 Potash, 17-09 33- Soda, 8-60 11 Lime, 17-31 27' Magnesia, . . . . 1-90 4- Ferric oxide, .... "33 Phosphoric acid, . . . 27-04 29- Sulphuric acid, ... Chlorine, .... 9-87 16- The chlorine is in combination with the alkalies, the iron and "the earths occur as phosphate, as well as the potassium oxide. So that the mineral constituents of cows' milk are, phosphate of potash, phosphate of lime and magnesia, common salt, and a trace of phosphate of iron. Other mineral inorganic constituents have been found in small quantity. If sufficient milk be used, it is not difficult to obtain a fluorine reaction, and since fluorides form :an essential constituent of the teeth, it is easy to see their impor- tance. A minute quantity of sulphuric acid as sulphates exists § lis.] THE COMPOSITION OF COWS' MILK. 247 in Biilk, averaging from -05 to -08 grm. per kilogramme; and it- has also been asserted by G. Mnsso, that milk contains a sulpho- cyanate. This assumption was based on the following experi- ment : — 15 litres of milk, freed from casein, fat, and albumen, •were neutralised by baryta water, and evaporated to a syrup, and the syrup extracted with absolute alcohol; the alcoholic extract dissolved in water and treated with zinc and sulphuric acid, yielded some hydric sulphide ; and subsequent treatment yielded from 6 to 21 mgrms. of barium sulphate per kilogramme.* The experiment a])pears to the author as hardly conclusive of the presence of a sulphocyanate, and reqiiires fui'ther investigation. Nadler has ascertained that neither cows' nor goats' milk ■contains any iodine : he used for the reseai'ch 6 litres of cows' milk and 3 of goats' milk.f Minute traces of copper have been found in milk ; but lead, arsenic, and all other metals, save iron, are absent. § 118. Other Constihients of Milk.— In 1864, E. Millon and Commaille, after coagulating and separating the casein and albumen, obtained a precipitate from the yellow whey by means of a solution of mercury nitrate. This precipitate was white, amorphous, and became slightly red on drying ; it was insoluble in water, alcohol, and ether. The precipitate was washed with water, then with alcohol, and finally with ether, and after drying weighed. Cows' milk yielded . . . 2*9 to 3-4 per litre. J Goats' milk ,, ... 1"52 ,, Sheeps' milk ,, . . . 2-53 ,. Asses' milk ,, ... 3-28 ,, Woman's miik,, . . , 2'77 ,, To this body they asciibed the following formula — CgoHg^I^TgOjg, HgO + HgO, NO3, and gave it the name of Lacto-proteine. In 1879, the author studied this body and decomposed it, and came to the conclusion that lacto-proteine, as a single definite substance, had no existence ; but that the mercuiy precipitate was composed of two substances to which the names of galactin and lactochrome respectively were ascribed. With these substances are precipitated small portions of albumen, which may have escaped precipitation, and traces of urea. The method of separation adopted by the author is as follows : * Berichte der Deutschen Cheinischen Oei^elhchaft, xi., p. 154, 1878. t Ueber den angebliehen lodgehalt der Luft und verschiedener Nahrungs- mittel. Journal fi'ir Praktische C/iemie, 99, p. 19S. J Nouvelle Substance contenue dans le Lait. Extrait d'une Note de MM. E. Millon et Commaille. Comptes Rendus, 59, p. 301, 1864. 248 foods: their composition and analysis. [§ 119. The casein and albumen are separated in the manner described at p. 282; the yellow whey is then precipitated by a solution of nitrat(! of mercury, of about the same streuyth as that used for estimation of urea;* the dense flocculent precipitate is tlien, after suitable washing, suspended in a A'ery little water, and decom- posed by hydrogen sulphide. The liquid is filtei-ed, and to the tiltrate a slight excess of acetate of lead in solution is added, when a dirty-white precipitate falls, which is collected, decom- posed by hydrogen sulphide, and recomposed by acetate of lead until it is obtained perfectly white. On combustion, numbers are obtained agreeing with the following formula, ll(PbO).,Pb02 Galactin, as obtained by decomposing the lead salt by hydrogen sulphide, presents the appearance of a white (or, if slightly im- pure, a fawn-coloured), brittle, neutral, tasteless, non-crystalline mass, soluble in water, insoluble in strong spirit. It presents some of the characters of a pei)tone, and can be separated from the yellow whey of milk by the general alkaloidal reagents of Sonnenschein and Scheibler.f After the galactin lias been removed from the liquid, an alkaloidal colouring- matter (for which the author proposed the name of lacto- chrorne) remains in solution, and may be pi*ecipitated by means of nitrate of mercury. The simplest formula for this appears to be HgOCgHjgNOg. Lactochrome, as obtained by careful decomposition of the mercury-compound, is in the form of bright i-ed, orange, resin-like masses, softening at 100°, freely soluble in water, very soluble in hot alcohol, but partially separat- ing as the liquid cools. There appears little doubt that lacto- chrome is the cause of the yellow colour of milk whey, and also the colouring-matter of butter. Thudichum described some years ago,'| under the name of butyro-luteine, the spectroscopic appear- ances of the colouring-matter of butter. § 119. Two bitter principles, possibly derived from substances eaten by the cow, were separated from the milk of Devonshire cows by the author in 1879. A commercial gallon of milk, measuring 3800 cc, was freed from casein, albumen, and alka- * The ordinary sohition for the estimation of urea is made by dissolving 100 grms. of pure mercui-y in half a litre of nitric acid ; a further quantity of acid is added until no red fumes are evolved; the solution is evaporated to a syrup, and after adding enoui,di nitric acid to prevent the formation of a basic salt, it is made up with distilled water to exactly 1400 cc, each cc. = "01 urea. t Sonnenschein's reagent is a nitric acid solution of phospho-molybdic acid. Scheibler's reagent is a solution of phospho-tungstic acid. X " Chemical Physiology," by J. L. W. Thudichum, M.D., London, 1S72, p. 149. § 120.] THE COMPOSITION OF COWS' MILK. 249 loidal matters by successive treatment in the way indicated at pp. 248 and 262, and the whey precipitated by solution of mercury nitrate, the mercury precipitate separated by tiltration, and the excess of the mercury nitrate thrown out by hydrogen sulphide; tlie liquid was then made alkaline by ammonia, and lastly pre- cipitated by tannin. The ])rt'cipitate was waslied, dried, and triturated with litharge and alcohol ; and, finally, the mixture was exhausted by boiling alcohol, filtered, and the filtrate evaporated to dryness. The dry residue was now dissolved in watei', digested with animal charcoal, filtered, and evaporated to dryness. A dark sticky extract was the result of these various and successive operations. On treatment by absolute alcohol, about 20 mgrms. of niimite white crystals were obtained, which, on combustion with cupric oxide, were found to be non-nitrogenous, and (so far as a small fpaantity can be trusted) gave a formula closely agreeing with CnHgOn,. The larger portion of the ex- tract was of a dark bi'own colour, very hygroscopic, and becom- ing quite fluid on exposure to the air; soluble in water in all proportions, insoluble in strong alcohol, and reducing copper solution on boiling, as well as chloride of gold at the ordinaiy temperature ; it also rapidly reduced nitrate of silver on warm- ing. The taste was woody and feebly bitter ; the reaction was neutral. Two strictly concordant analyses gave numbers agree- ing with the formvila Ci8Hi80;;4. Kreatinine appears to exist in normal milk, for the author has separated, in the ordinary way, -059 of the chloride of zinc compound of kreatinine from .310 cc. of milk; this corre- sponds to -0115 of kreatinine in 100 cc. of milk. Commaille had already discovered that in a sample of milk which remained some time in a partially closed flask, appropriate treatment would give ciystals yielding with chloride of zinc, or oxide of mercury, kreatinine reactions ; and the presence of this substance in small quantities as a normal (perhaps constant) constituent of milk is now placed beyond a doubt. Urea has been found in almost all animal fluids, and in such small quantities as a milligramme per 100 cc, it is rarely absent, but any marked percentage of urea is, of course, abnormal. § 120. Milk has a peculiar sweet odoui-, and the odoriferous principle may be separated by agitation with three or four times its volume of petroleum ether. MINI. Millon and Commaille have recommended bisulphide of carbon for this purpose, but it is not quite so convenient a solvent.* In neither case is any of the milk-fat dissolved. Lactic acid is almost invariably present in * Analyse du Lait, MM. ^Millon et Commaille. Compt. liencL, t. 59, 396. 250 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 121. cows' milk, and gradually increases in quantity. The amount of lactic acid obtained from milk which has been drawn from the cow within the hour is from -01 to -02 per cent. Citric acid is also a normal constituent of milk. § 121. Putting in one view, the constituents of milk, so far as is known at the present time, they are as follows : — AVERAGE COMPOSITION OF HEALTHY COWS' MILK. Parts per cent. by weight. f Olein, , 1-477 ^ 1 Stearin and . 1 Palmitin, • j 1-750 1 1 Milk-fat . ■{ Butyrin, 1 Caproin, 1 Caprylin and L Rutin, . 0-270 '. (oO03 1 1 J 3-50 Casein, . 3-98 Albumen, . . 0-77 Milk-susar, .... 4 00 Peptone-like body (galactin), . . 017 Lactochrome, ........ undetermined Amorp^hous, bitter principle (gWide ^ | p7,^4Sd j 0-OP ^'^ ^^ ^*' ( by tannin, ) Crystalline principle, CoHgOio, .... undetermined Citric acid,+ from '09 to -11 Lactic acid, . . , Absent in milk in the udder, but by the time an analysis can be made, always present from '01 per cent. Alcohol, traces always present Odorous principle, oil of milk? .... undetermined Urea, .... traces, such as -0001 per ceut., nearly always present. Kreatinine, , traces (Commaille). Ash . I K„0, I Xa„0, I CaO, ^ Fe.Oj I P.O., I CI, UigO, 0-1228^ 0-0868 I 0-1608 i 0005 1^ 01922 I 0-1146 I 0-0243 J 0-70 Fluorine, .... Sulphuric acid in combination, Suljjhocyanates, . Water, .... very minute traces. •005 * Mean of four determinations only. t Soxhlet, Chem. Centr., 1888, 1067-1068. 86-87 122.1 GASES OF MILK. 251 GASES OF MILK. § 122. Tlie author has investigated the gases contained in milk. Various samples of milk were clamjjed on to a mer- curial pump, and the whole of the gas which they yielded pumped out and received in tubes, whence the gas was trans- feri-ed to a gas apparatus and analysed. A litre of new milk, while fresh and warm from the cow, connected in this way to the pump, yielded 1-83 cc. of gas, which on analysis had the following composition : — cc. Per cent. Carbon dioxide, C0„, -06 . . . .3-27 Nitrogen, N, . ." 1-42 . . . 77-60 Oxygen, 0, . . -35 . . ,19-13 The proportion of oxygen to nitrogen was therefore nearly as 1:4. Another litre of good Devon commercial milk, on being subjected to the same process, yielded 3*468 cc. of gas, the percentage composition of which was : — Per cent. Carbon dioxide, CO 2, . . . . 60-47 Nitrogen, N, 30-21 Oxygen, 0, 9-30 This sample had been standing at a temperature of 15° for some hours ; hence the diminution of oxygen and the increase of carbon dioxide. Various other similar experiments were made, with the result of establishing the fact that a litre of fresh milk yields to the Sprengel pump from 1 to 3 cc. of gas, in which there is always a certain percentage of carbon dioxide, and in which the relation of the nitrogen to the oxygen is very similar to the relation that exists in the air dissolved in water ; but that fermentation, at any temperature in which fermentation is possible, at once commences, when the lactic ferment begins to use up the oxygen, and ultimately carbon dioxide is the only gas which can be obtained. This evolution of cai'bon dioxide is slow but continuous. As an example of this fermentation, one of the experiments may be cited : — 100 cc. of milk in which fermentation had begun, were suitably clamped to a Sprengel pump : on the first day there was a small percentage of nitrogen and a little oxygen ; and on the second day a trace of oxygen ; but on the succeeding days the gas consisted wholly of carbon dioxide, as follows : — 152 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 122. 1st (lay, 2nd day, 8rd day, 4th day, 5th day, 6th day, 7th day, 15th day, 19th day. 1-123 cc. ofCOa 5-086 ,, ,, 19-540 ,, „ 7-621 „ „ 7-370 ,, ,, 9-023 „ ,, 1-780 „ ,, 21-350 „ „ 4-370 ,, Giving a total of 77'263 cc. of carbon dioxide in 19 days, the temperature ranging between 14° and 19°-5. Milk previously deprived of dissolved air by the Sprengel pump, then confined over mercury and submitted to an atmosphere of oxygen, rapidly absorbs the oxygen, the place of wliich is taken by carbon dioxide, provided the temperature is a fermentation one — that is, above 9° and below 60°. This continuous absoi-ption of oxygen was well shown in an experiment of the author's, in which a litre of milk was submitted to the action of a Sprengel pump, and in which it was found there was a continuous slow diffusion of air through the india-rubber connections. It has long been shown by Graham, that iiir thus finding its way through the minute pores of thick rubber is very highly oxygenised ; yet all oxygen rapidly disappeared from the gas, and after the second day pure nitrogen and carbon dioxide could alone be obtained : — 1st day, 6-732 cc. of gas. Percentage composition. Carbon dioxide, 55-392 Nitrogen, 33-780 Oxygen, 10-828 Eatio of oxygen to nitrogen as 1 to 3 nearly. 2nd day, total gas, . . , 7 2 cc. Carbon dioxide, . . . . , . 40-73 Nitrogen, ....... 49 73 Oxygen, -54 3rd day, 4-803 cc. Carbon dioxide, 61-06 ISitrogen, 38-94 4tli day, — Carbon dioxide, 87-98 Nitrogen, 12 02 § 123.] FORE MILK. 253 Percentage composition. 5th day, — Carbon dioxide, Nitrogen, , 91-52 8-48 "FORE" MILK. § 123. If an animal is fractionally milked — that is, the whole of the milk received into three or four different vessels — it will be found that, on analysis, the several portions exhibit some difference of comi)osition, more observable in the last and the first, than in the intermediate portions. This difference mainly affects the fat, the first portions of the milk yielding, as a rule, but little fat, while the latter portions, called " strippings " (in speaking of cows' milk), contain an excess of milk-ftit. Thus, in a Devon cow milked in this way for the purpose of analysis, the writer found the two extreme portions to have the following composition : — Specific gravity. Milk-fat, . Casein, Aibumen, . Peptones, . Milk-sugar, Asb, Water, Common salt in ash, Fore Milk. Strippinss. 1-0288 1-0256 1-16G 5-810 2-387 4-304 1-S30 •975 -381 -545 3-1-20 3-531 -797 -895 90-319 83-940 In another experiment a Guernsey cow yielded the following ; Specific gravity. Milk-fat, . Casein, Albumen, Peptones, Milk-sugar, Ash, Water, Common salt in ash, Strippings 1023 5-946 3-435 •860 •156 5-280 •929 83-394 •098 Dairymen are perfectly aware of the poorness of fore milk in fat, and more than once fraudulent milkmen have endeavoured 254 foods: their composition and analysis. [§ 123. to defend themselves by having recovii'se to the strange expedient of partially milking a cow before such functionaries as aldermen or policemen, and delivering with all formalities the sample to be analysed. The analyst, not knowing its history (for in such cases it is transmitted as an ordinary commercial milk), and finding it on analysis deficient in fat, certifies accordingly, and until the matter is explained suff'ers in I'eputation. Such tricks have during the last few years been rather common, but so fully exposed that they are not likely to re-occur. This difierence in the first and last milkings is not confined to cow.s' milk, but has also been observed in the milk of other animals. Peligot had an ass milked in three successive portions, and found as follows : — 1 2 3 Milk-fat, ... -96 102 1-52 Milk-su^ar, . . . 6-50 6-48 6-50 Caseiu,' .... 1-76 1-95 2-95 TotaUolids, . . 9-22 9-45 10-97 Water 90-78 90 oS 89 03 Reiset* has also found a considerable difierence in the per- centao-e of total solids in human milk in fractions taken before the child was applied to the breast and after. Before Suckling. After Suckling. Total Solids per cent. Total Solids per cent. 1 10-58 12-93 o .... 12-78 15-52 3 13-46 14-57 It has been considered that this diff'erence is merely due to the efiect of a physical cau^ ; that, in short, as regards cows' milk, the milk already secreted is in the same state as if it stood in a vessel, and the fat rising to the top is, of course, drawn last. This explanation cannot be altogether true, for the same phenomenon is observed in human milk, and here the breasts are horizontal, or nearly so. It is more probable that during the act of milking secretion goes on, and it would seem that the fatty contents of the milk-producing cells are set free before the more watery and albuminous. Hence, the strippings are, as the most recent portions of the whole secretion, rich in fat. This view is supported by an experiment of Eeiset, in which it was proved that the longer the time elapsing between the partial milkings, the less the percentage of solids. * J. Reiset : Annates de CJiimie et de PInjsique, 3 ser., xxv., 1849. § 124.] HUMAN MILK. 255 HUMAN MILK. § 124. Woman's milk has been long an object of research, and numerous analyses of it are scattered through scientific literature. These analyses, in their quantitative results, show considerable dis- crepancies, so that we must either adopt the supposition that human milk is very variable, or, what is more probable, that the samples taken did not represent the average secretion. From experiment, the author has come to the conclusion that it is impracticable with any mechanical appliances to obtain a complete sample of human milk. In civilised life the nervous system assumes such a high and delicate state of organisation, that the secretion is far more dependent on the presence and contact of the oflspring than among animals. Hence, samples of human milk taken by bi-east pumps, or other exhaust apparatus, can only be considered partial samples ; and a study of partial sampling in the case of cows' milk (p. 253) has taught us how veiy widely the quantities of the fatty constituents in such samples differ from one taken from the whole bulk. Woman's milk contains milk-fat which has not been obtained in quantities sufficient for accurate investigation, and a knowledge of its exact composition is still a desideratum. It, however, certainly contains butyrin, for the author has succeeded in isolating a sufficient quantity of butyric acid from saponified human milk- fat to identify it satisfactorily. Milk-sugar, casein, albumen, peptones, and a colouring-matter, with mineral substances, are also constituents of woman's milk.* The casein, like that of the ass, is peculiar in not separating in flocculent masses by the pi'o- cesses recommended at p. 282), and the analyst is under the necessity of adopting a different process. This difference is all- important ; for in artificial feeding with cows' milk, as soon as the milk reaches the stomach, the milk, in popular language, " curdles," and is often rejected by vomiting. One of the earliest exact analysis of human milk was made by Meggen- hofent in 1S2G. His treatise scarcely appears to be known, yet it contains pretty well all that is known of the composition of human milk. The total solids of human milk Meggeuhofen determined from twelve samples, the highest of which is 13-38, and the lowest 9 25 per cent., the mean being about 12 per cent. Probably for the first time Meggeuhofen determined the * Human milk on being shaken up with ether parts with its fat, the globules dissolving in the ether, the liuid therefore clears up and separates into two layers — an upper ethereal containing the fat in solution, and a lower layer consisting of a solution of the casein, albumen, and salts. This pecu- liarity is not shared by the milk of the herbivora. Hue P. Radenhausen,, Zeit.f. phyfiioloijUcht (Jhtmit, 5, 13-3U. ■\- Dinsertatio Inauguralis Indar/ationem Lactis Muliebris Chemicam. C. Aug. Meggenhofen. Frankfort, IS'2{J. 25G foods: their composition and analysis. [§ 125. albumen separately from the casein, and also weighed " inaferice animalis tiiictura gaUarnm animalis jn-ecipitatce." His view of the composition of human milk may be fairly stated thus — Per cent. Milk-fat, 2-90 Casein, 2-40 Albumen, '57 Albuminoid precipitated by gallic acid, . . "10 Suaar, 5-87 Ash, -16 Water, 88-00 According to tlie writer's own experiments on human milk, iind the quantitative analysis of samples taken as fairly and completely as can be done, it has the following composition : — Milk-fat, . . . . 2-90 Casein, . . . . 2-40 Albumen, . . . . •57 Peptone (galactin). •10 Su-ar, . . . . . 5-87 Ash •16 Water, . . . . . 88-00 Total solids, 12 Solids not fat, 91 "With regard to other constituents, urea is often present ; there is also an odorous principle. Human milk decomposes similarly to cows' milk, and yields similar gaseous and other products.* § 125. 31ilk of the Ass. — The author has investigated the milk of the ass. Milk was obtained under his personal superinten- dence from asses kept and fed in London dairies for the purpose of supplying the demand that still exists for asses' milk. The * A. R. Leeds (Chem. News, 4, 263-267, 250, 281) has examined eighty- four samples of human milk, and gives the results as follows : — Specific gravity, Albuminoids, . Su^ar, Fat, Ash, Water, . Solids not fat, . '1 otal solids, . All the samples were alkaline in reaction, 10353 1-995 4-86 6-936 7-92 4-131 6-89 -201 •37 86-732 89-08 9-1.37 12-09 13-267 16-66 1 0268 •85 5-40 211 0^13 83 21 0^57 10-91 § 120.] MILK OF THE GOAT. 257 animals were fed on a uniform diet of bran, hay, and oats. The yield of each milking was carefully noted, and the ass in each case milked dry. It would appear that the milk of the ass under these circumstances has a very imiform composition, the difierences observed being quite unimportant. The yield for commercial purposes appears not to exceed 3 pints, and to aver- age about 2h pints daily. More than this is doubtless secreted, but some of it is used by the foal. In no case did a single milk- ing yield half a litre (three-fifths of a pint), but usually between SOU and 400 cc. The mean composition of asses' milk is as follows : — Per cent. Milk-fat, 1-02 Casein, 1-09 Albumen, ........ '70 Peptones, -10 Sugar, 5-50 Ash, -42 Water, 91-17 Total solids, .... S-S3 Solids not fat, .... 7*81 The fat contains 5 per cent, of butyric acid, equal to 6-6 per cent, of butyrin ; it is probably very similar to butter-fat. After the precipitation of casein, albumen, lactochrome, and peptones, there yet remain principles precipitable by tannin. As in human milk, the casein is not readily precipitated, but remains suspended in a state of tine division, however far lactic fermentation may have progressed. § 126. Milk of the Goat. — The milk of the goat,* as a rule, contains more cream than that of the cow, and rather less albuminous matter. * Some analyses by Voeloker {Bied. Centr., 1881, 858) of goats' milk give the following results:— Water, , Fat, _ . Casein, . ; Milk-sugar, . Ash, . Solids not fat, Specific gravity, Per cent. Rich. Poor. Colostrum 82-02 84-48 83-51 7-02 6-11 7-34 4-67 3-94 3-19 5-28 4 -68 5-19 101 -79 -77 [10-96] 1-0357 [9-41] 1-030 18 258 FOODS : THEIR COMPOSITION AND ANALYSIS. [§12- Its average composition is as follows Per cent. Milk-fat 4-20 Casein, 300 Albumen, . •62 Peptones, . •08 Milk-sugar, . . . . 400 Mineral constituents. •56 Water, . 87-54 Total solids. 12-46 Solids not fat, . 8-26 § 127. Milk of the Mare. — The milk of the mare closely re- sembles in its constituents the milk of the cow : the casein, the sugar, and the fat being veiy similar, if not identical. M. J. Duval* asserts that he has discovered in the milk of the mare a new acid, to which he has given the name of equinic, and which crystallises in groups of little needles ; it is not volatile without decomposition, in odour fragrant. It is combined with a base volatilised by heat, which the author considers a base of the ammonia type. Its reactions with silver nitrate, ferric chloride and auric chloride distinguish it from hippuric acid. No analyses are, however, given. The mean composition of mare's milk is as follows : — Milk-fat, 2-50 Casein, 2-19 Albumen, "42 Peptones, '09 Sugar, ......... 5 '50 Mineral constituents, ...... '50 Water, 88-80 Total solids, . . . . 11-2 Solids not fat, ' . . . . 8-7 Comptes Rendus, t. 82, 419, 1876. §§ 128-131.] MILK OF OTHER MAMMALS, ETC. 259 MILK OF OTHER MA]MMALS ; LACTESCENT PRODUCTS OF BIRDS AND PLANTS. The following Notes on the Composition of the Milk of other Mammals, &c., may be found useful for comparative purposes. § 128. Milk of the Sheep. — Sheeps' milk* is remarkable for its high specific gravity, aud the large amount of solid matter which it contains ; the specific gravity ranges from lOoS to 1011, and the total solids may rise as high as 19 per cent. The average composition is as follows : — Per cent. Milk-fat, 5-30 Casein, ........ 6 "10 Albumen, I'OO Peptones, 0'13 Milk-sugar, 4 "20 Ash 1-00 Water, 82 -27 Total solids, . . . . 17-73 Solids not fat, .... 12-43 The casein behaves similarly to the casern of cows" milk, and separates easily by dilution, acidulation with acetic acid, &c. (see p. 282). The fat yields 5 per cent, of its weight of butyric acid, and is probably of similar composition to the milk-fat from cows' milk. § 129. The Milk of the Camel. — Chatin has analysed the milk of the camel. Hedescribesfitasperfectly white in colour, and possessing globules smaller but more numerous than those in cows' milk, the diameter being on an average one-half. Specific gravity, 1 042. It appears to be rather richer in milk- sugar and casein than cows' milk. Drageudorfi'J has also analysed camels' milk, and gives the following figures : — Per cent. Albuminoids, 3-84 Fat, 2-90 Milk-sugar, 5-66 Ash, -66 Water, 86-94 § 130. Milk of the Llama. — Doy4re§ has analysed the milk of the llama. The mean of his three analyses is as follows : — Per cent. Milk-fat, ........ 3-15 Albuminoids, -90 Milk-sugar 5 60 Ash, -SO Water 89-55 §131. Milk of the Hippopotamus. — There are few opportunities of analysing * An analysis, by Voelcker, of rich samples of ewes' milk shows that the fat may attain 12-78 per cent. Bied. Centr., 1881, 858. •\ Sur le Lait de la Chamelle a deux Bosses, par M. Chatin. Journal de Pharmacie et C'hemie, t. i., 4 ser., p. 264. J Zeit.f. Chemie, 1865, s, 735. § Ann. de VInst. Agrom. 1852, p. 251. 2G0 foods: their composition and analysis. [,§.^132,133. the milk of this enormous animal, as it is fierce when it suckles its offspring. A sample of the milk was, however, investigated by Gunning." He describeslt as of an acid reaction, and under the microscope showing larger globules than that of other animals. The young hippopotamus sucks under water, and can remain there for a much longer time than the adult animal. The secretion of milk is excessive in quantity, and escapes from the distended teats in streams, which make the water around the animal quite opaque. Its general composition appears to be as follows : — Per cent. Milk-fat, 4-51 Milk-sugar, with a small portion of albuminoid substance, ....... 4 '40 Salts, -11 Water, 90-98 § 132. Milk of the Sow. — The mean of eight analyses collected by Konigf of sow's milk, is as follows : — Per cent. Milk-fat, 4-55 Albuminoids, 7 '23 Milk-sugar, 3-13 Ash, 1-05 Water, 84-04 Total solids, 15-9G Solids not fat, 11-41 There are also two analyses of the milk of a sow investigated by Filhol and Joly ; the animal was fed on horse-llesh, a diet far from natural ; under this diet -was secreted a highly albuminous tiuid, containing but little sugar. Specific gravity 1-044. Per cent. Albumen, 12-89 21*0 Fat 6-6 5-4 Sugar, 0-5 12 Extractives and salts, .... 3-01 43 Water, 77-0 68-1 Total solids, . . . . 23-0 31-9 Solids not fat 16-4 26-5 § 133. Milk of the Bitch.— The milk of the bitch is highly charged with albuminous solids, and is of a specific gravity ranging from 1-0.34 to 1 -036. It has been investigated by Simon, Dumas, Filhol and Joly, Talmatescheff, Bensch, Scubotin, and others, with the following mean results : — Per cent. Milk-fat 9-57 Casein, 5 "53 Albumen, 438 Milk-sugar, 3-19 Ash, , . • -73 Water 7660 Total solids, 23 40 Solids not fat, 13-83 * Gazetta Chim. Italiana, 1871, p. 255. t Op. cit. §§ 134, 135.J MILK OF OTHER MAMMALS, ETC. 261 § 134. Milk of the Cat. — The milk of the carnivora generally has the pecu- liarity of having the milk-sugar almost entirely replaced by lactic acid, and hence the milk invariably possesses an acid reaction. An analysis of the milk of a cat by Commaille is as follows. The milk was taken twenty-four hours after kittening ; it was feebly acid : — Pftr cent. Milk-fat, H-333 t'asein, . . . . . . . . 3' 117 Albumen, ........ B'Ooi Lacto-proteine, ....... '407 Lactose and organic acids, . . . . . 4 '9 1 1 Ash, ......... '560 Water, 81-GJ3 Total solids 18-377 Solids not fat, 15-044 The -4G7 would correspond to about '25 of peptone. § 1.35. Milh-like Secretions of Birds and Pla)its.— It is usually held that mammals alone secrete milk, but this is by no means certain ; for during the latter portion of the incubation-period, as well as more ])rofusely for a little while after the young birds are hatched, the pigeon secretes a nutritious albuminous fluid in her crop, which is sujjposed to be used for the purpose of feeding the young birds. According to Lecomte's analysis this secretion contains, Casein and salts, 2.>'23 Fat, 10-47 Water, GG-30 Such milk-like secretions are by no means confined to the internal mucous membranes of birds. Jonge* has made a most valuable research on the secretion of the glands known to anatomists as GCcuniukt uropyfjit, situated at the tail of the common goose. The secretion was obtained in sufficient quantity for a comj)lete qualitative and quantitative analysis, and although the analysis was not quite so complete as if a larger quantity had been obtainable, it fairly shows that there is a considerable analogy between milk and this secretion, the most marked difference being that no trace of milk-sugar could be found. The analyses of two samples were as follows: — 1. 2. Total solids, . . . 391-03 415-.34 Water, . . . . ^ 608-07 _ :84GG Albuuien and nuclein, . 179-66 127-63 Compounds insoluble iu absolute ether, . . 186-77 247-08 Alcoholic extract, . . 10-9!) 18-31 Water extract, . . 7 03 11-31 Ash 7-07.^'^«^- -"^''M 11-07 ^^"1- '-71 ^^''' • • • • ' "' ( Insol. 3-36 1 '-^^' llnsol. 3-36 * " Ueber das Secret der Talgdri'isen der Vcigel und sein Verhiiltniss zu den fetthaltigen Hautsecreten der Saugthiere, insbesondere der Milch," Von D. de Jonge. Zeitschri/t Jar physiol. Chemie, Von F. Hoppe-Seyler. Strasburg, 1879. 262 foods: theiu cOxMposition amd analysis. [§ 1-36. In ether extract. 1 2 Cetyl-alcoliol, 74-23 104-02 Oleic acid, 6-48 Lower acids, . 3-73 14-84 Lecithin, 2-33 In the vegetable kingdom, numbers of trees or plants yield a white fatty secretion, popularly called milk, though, as a rule, such fluids have no right to this title, being totally different in composition and ])roperfcies. A very remarkable exception to this assertion is, however, met with in the "milk tree" (Brosimum galaclodenilron). to l)e found in Central America. This tree, on incision, yields an abundance of a thickish feebly acid fluid, coagulating on exposure to the air. ]\I. Boussiugault has recently analysed this juice, and considers it perfectly analogous to ordinary milk, since it contains a fatty ])rinciple, an albuminous principle, a sugar, and phosphates. The exact composition of these different matters has, however, not been determined. Boussingault's general analysis is as follows : — Fatty saponifialde matters, . . . . 35-2 Sugar, and substances analogous, . . . 2-8 Casein, albumen, . . . . . 1-7 Earths, alkalies, phosphates, ... -5 Substances not estimated, . . . . 1-8 Water, 58-0 This milk is used largely as a food in the i-egions where the tree grows. ABNORMAL MILKS. § 136. Milk which deviates from the natural secretion, the animal suffering from no disease, and milk secreted under unnatural conditions, may be conveniently classed as " abnor- mal." (Milk derived from the unhealthy will be considered in another section.) Instances of ItealtJiy cows giving milk dif- fering essentially from ordinary milk are very few. One such, however, is recorded by Mr. Pattinson, who analysed the milk of a roan cow, which only gave 2 per cent, of albuminoids, and yielded no less than 4 grms. per litre of common salt. The animal is stated to have been in good health. Instances of marked deviations from the ordinary standard are to be found in The Analyst of Jan. 1, 1893. The newly-born human infant almost constantly secretes a fluid in the mammae, and adult males have not only secreted milk, but that in abun- dance enough to suckle. Females also, both human and animal, occasionally secrete milk without having been previously pregnant. With regard to the milk secreted by infants, there is some doubt about its real nature. KiJlliker does not view it as a true milk, but considers its appearance connected with the formation of the mammary glands. Sinety, on the other hand, upon ana- tomical grounds, considers it a true lacteal secretion. It probably is a sort of imperfect milk loaded with leucocytes, and this is the more likely, as Billard (Tra'dii des Maladies des En/ants nouveau nes, 3me edition, 1837, p. 717) notices that it frequently ends in abscess. Schlossberger gives an imperfect quantitative analysis of a sample of § 136.] ABNORMAL MILKS. 263 milk,* obtained by squeezing tlie breasts of a newly-born infant — a male. In the course of a few days about a drachm was obtained. The following was the result of the analysis : - Per cent. Water, 96-75 Fat, -82 Ash, -05 Casein, sugar, and extractives, . . . 2 '83 Sugar reaction strong. The most complete analysis we yet possess of such milk is one by V. Gesner, which is given in the following table with other less perfect analyses : — It 2i 3§ Milk-fat, . . . 1-456 '82 1-40 Casein, .... '557 ... ) r, ^n Albumen, . . . -490 ... \ " *" Milk-sugar, . . . •956 ... | ^ .^ Ash, .... -8-26 -05 i •'*" Water, .... 95705 96-30 3940 Total solids, . . . 4-295 3*70 10-60 Joly and Filhol have recorded the case of an old lady, 75 years of age, who suckled successfully her grandchild. || Similar instances have been i-ecorded in dogs, and we fortunately possess one or two anal3'ses which show that the fluid is certainly milk. Thus Filhol and Jolly give the following analysis of the milk derived from a bitch which had no comiection with a male : — Specific gravity, ...... 1 -069 Total solids, 29 00 Fat, 2-20 Sugar, -32 Albumen, 23-20 The ash, on analysis, gave the following i)ercentages in 100 parts : — Chloi-ide of sodium, ..... 6o-10 Chloride of potassium, . . . . 3 -SB Calcic phosphate, 27'75 Sodic phosphate, ..... 1-40 Sodic carbonate, I'Sl Tj-aces of maguesic and otlier ])hosphates. Men before now have suckled children. HumboldtU relates the case of Francisco Lozano, whom he saw, and whose case he carefully investigated ; and it appears established that this man did secrete from his breasts a nutrient tluid on which his infant son lived for many months, it is said, indeed, a whole year. The curious in such matters may consult the refer- ences given in the footnote for additional cases.** * Untersuchung der sogenannte Hexenmilch, J. Scblossberger, Annaleii der Chemie u. Phurmacie, b. 87, 1852. ■\ Jahrh.f. Kmderhrankheiten, N. F., Bd. ix., §160. J Scblossberger u. Hauff, Ann. Chein. Pliarm., Bd. xcvj., p. 68. § Gabler u. Quevenne, op cit. II "Recherches sur le Lait," iii., Bruxelles, 1856. "' Humboldt: "Voyage aux Regions Equinoxiales du Nouveau Continent." ** Robert, Bishop of Cork : Letter caucerning a Man who gave Suck to a Child, Phil. Trans., 1741, No. 461, t. xli., p. 813. Franklyn : "Narrative 264 foods: their composition and analysis. [§137. Instances have also been known of a like kind among animals. Schloss- berger has analysed the milk derived from a he- goat (Annalen der C'hemie 11. Pharmacie, 1844) : Milk-fat, 26-50 Casein, with salts soluble in alcohol, . 9 'GO Sugar, with salts soluble in alcohol, . . 2 'GO The ash was -782 per cent. — viz., -.325 soluble in water, -457 insoluble. Occasionally the female mammae after continement have continued to yield milk, although the infant has either been dead or nourished otherwise. In such cases the milk deviates from its normal composition, and is, for the most part, highly albuminous. In a case of this kind recorded by Filholand Joly,* three analyses of the milk were made as follows, at different dates, about a week apart : — 1. 2. 3. Specific gravity, . . . 1-039 1-025 l-0'23 Total residue, . . . 21-50 1S-.30 18-G3 Milk-fat, .... 5-00 G15 7-80 Sugar, 2-19 127 3-50 Albumen, . . . .12-96 9 00 5 65 Extractives and salts, . . 1 -35 1 -88 1 68 Water 78 50 81-70 81-37 Casein was entirely absent. The composition of the ash was as follows : — Per cent. Chloride of sodium, . . . . . . . 73-10 Cidoride of potassium, traces. Calcic phosphate, . ...... 23-40 Sodic phosphate, -SO Sodic carbonate, ....... 1-89 Magnesic and ferric phosphates, .... -81 GENERAL EXAMINATION AND ANALYSIS OF MILK. § 137. The general examination and analysis of milk may be conveniently treated of under the following heads: — I. Microscopical and biological e.xamination of milk. II. Analytical processes, more particularly for the purposes of the food-analyst. A. General analysis of milk. {a.) Specific gravity. (6.) Total solids, (c.) Extraction of milk-fat. (rf.) Extraction of milk-sugar, (e.) Albuminoids and ash. of a Journey to the Shores of the Polar Sea," 1819, p. 157. Cobbold : Milk from the Male Mamma, ilonlhlij Journal of Med. Science, 1854 ; t. xviii., p. 271. Morgagni : Adveisaria Anatomka Omnia (T. Animad- versio, i., p. 3). * Comptes Eendus, t. xxxvi., jt. 571. 1853. § 13S.] EXAMINATION AND ANALYSIS OF MILK. 265 B. Various methods proposed for extracting the milk-fat. («.) Addition of sand, or sulphate of lime or other powder; (6.) Adam's method; (c.) Estimation of inilk-fat by centrifugal machines; (d.) Soxhlet's process; (e.) The Werner-Schmidt process of tat estimation. C. Various other methods of milk analysis. (1.) Drying in a vacuum. (2.) Direct determination of the Avater. (3.) Absorption of water by dehydrating agents. (4.) E-itthausen's copper process. (5.) Miiller's process. (G.) Clausnizer and A. Mayer's process. III. Special details as to the more exhaustive and scientific analysis of milk. (1.) Analysis of the milk-fat, and examination of the ethereal extract. (2.) Extraction of the milk-sugar. (3.) The ash. (4.) Estimation of albumen. (5.) Isolation of galactin. (li.) Estimation of the nitrogen of milk. (7.) Isolation of the principles precipitated by tannin. (8.) Estimation of urea. (9.) Estimation of alcohol. (10.) Volatile acids. (11.) Estimation of the total acidity of milk, and estimation of lactic acid. (12.-) Detection of metals in milk. (13.) Detection of nitrates in milk. I. Microscopical and Biological E.xamination of Milk. § 138. A mere chemical analy.sis is incomplete and insufficient in itself, and should in all cases be preceded or supplemented by a careful and painstaking microscopical examination. Normal milk, viewed under the microscope, presents for the most part a multitude of fat globules floating in a clear fluid. The globules of human milk measure in diameter from -002 to "005 mm.; those in the milk of the cow, from -00062 to -00039 inch [-0016 to -01 mm.] These fat globules are of two kinds. By far the most numerous are evidently drops of fluid fat; but there are occasion- ally to be seen others which would appear to consist of solid fat, for they are rougher on the surface, and less soluble in ether, characteristics which they lose on warming, becoming like liquid-fat globules. In human milk, and, to a certain extent, in cows' milk, there are also as normal constituents, but in sparse quantity — 266 FOODS : their composition and analysis. [§ 138. (1.) Fatty drops having a half moou-shaped, finely granular substance ; (2.) Clear cells enclosing one or two fatty drops, and an eccentric nucleus ; (3.) Round clear bodies, easily coloured by eosin and picro- carmine. These last Heidenhain considers to be free nuclei.* In the colostrum, or milk drawn the first few days after parturition, there are present other elements — viz., the so- called "colostrum cells." Some of these consist of a number of small and large fat globules, held together by a hyaline tissue or membrane, swelling on the addition of acetic acid or alkalies, and only slowly coloured by aniline red. There are other granular cells coloured at once by the same reagent. If the milk is taken fresh and warm, and a minute drop examined on a Strecker's warm stage,t and kept at a temperature of 38°, the coi'puscles will exhibit amoeboid movements, perfectly similar to those which liave been noticed in the white corpuscles of the blood. Indeed, it is almost certain that the colostrum cells are no other than the white corpuscles of the blood, infiltrated ■with milk-fat, for Heidenhain, having injected into the dorsal lymph vessel of the frog a cc. of fresh milk, after 48 hours found the white corpuscles loaded with milk-fat, and in no respect dis- tinguishable from colostrum cells. When the milk has undergone any fermentation, the lactic ferment itself may be identified, and little lumps of casein may be seen. These are mostly irregular and amorphous, but sometimes they have the appearance of flattened cylinders, and other shapes. In abnormal milk may be detected pus or blood, or sometimes both. If the pus is derived from inflammations within the mammpe, and has been mixed with the milk before milking, the pus cells become infiltrated with milk-fat, and are diflicult to distinguish from colostrum granules; but if derived from ulcers on the teats, they have the usual apjiearance of pus cells. The pus cells, like the colostrum cells, ami th^ mucus corpuscles, are all diflerent forms of white blood-corpuscles [leucocytes], and -when placed on the wai-m stage exhibit amoeboid movements. Pus cells, as usually observed, are s])heroidal, granular, and colourless, measuring from about 1-2500 to l-3000th of an inch in diameter. On treat- ment with dilute acetic acid, the cells clear up, and show two, three, or four nuclei. Blood, in small quantity, gives a pinkish Pl. Heidenhain: " Handbuch der Physiologic." Herausgegeben von Dr. L. Hermann. Leipzig, 18S0. t In defanlt of Strecker s stage, a plate of copper, having a central aper- ture and a thick straiglit wire, some inches in length, may be used. The plate is kept at the desired temperature through heating the wire by means of a spirit lamp. § 139.] EXAMINATION AND ANALYSIS OF MILK. 267 colour to milk; if a large amount be present, it sinks to the bottom in red flocculent mas.ses, which soon, from being deoxidised by" the millv, acquire a tint varying from a red more or less dark, to a shade almost black. In small quantities reliance must be placed on the microscopic appearance of the blood-discs, which are wholly unlike any cell found in normal milk. The red blood- discs of the cow are like those of the human subject — little circulai', biconcave, flattened discs, measuring on an average 1 -4000th part of an inch. Human blood-discs have an average diameter of l-3500th inch. By the aid of the micro-spectroscope, the absorption-bands may also be seen. These are, in oxidised blood, two bands between D and E, the one close upon the red being narrower, darker, and better defined than the one nearer to the green ; with deoxidised blood, only one baud is seen, between D and E. On treating the blood with oxygen, or shaking it up with air, the two bands re-appear. In "blue" inilks a peculiar fungus has been discovered, and in the milk from animals suff"ering from foot-and-mouth disease, certain special appearances have also been noted, which are described in the section treating of this disease. The bacteriological examination of milk is effected by culti- vating small quantities of milk in various media. The Bacillus coli has been found in many milks ; evidence of the tubercle bacilli has been obtained by feeding guinea pigs on milk and injecting milk into the subcutaneous tissue of the same animals. II. — Analytical Processes more Particularly for the Purposes of the Food-Analyst. § 139. One of the first accurate processes for the general analysis of milk was published in 1853, by MM. Vernois and A. BecquereL* A small quantity [30 grms.] was taken, dried, exhausted with ether, burnt up to an ash; the sugar obtained "a saccharimetre" from the whey, the casein being first separated by coagulation by acetic acid, and then estimated by difference. Mr. Wanklyn, by his work on milk analysis,! revived the more accurate method of using comparatively small quantities for analysis, thus avoiding very considerable error, from the risk of large quantities decomposing by prolonged heating. He advocated the use of platinum dishes, and suj^ported strongly the * Comptes Be)idus, t. ?>Q, p. 1S7, 1S5.3. t "Milk- Analysis." By J. A. Wanklyn. Loud. 1874. 268 FOODS : their composition and analysis. [§ 140. doctrine of the fairly constant character of the non-fatty consti- tuents of milk, dividing the total milk solids into two divisions : the one, ^^ milk-faf," the other, ^^ solids not fat." Whatever modifications have been since introduced in the methods for the analysis of milk, the general process is still on the principles advocated by Mr. Wanklyn. A. General Analysis of Milk. § 140. By the general analysis of milk is meant merely separa- tion of the milk, by the aid of solvents, into milk-fat, solids not fit, and ash. Such an analysis is the simplest quantitative exercise in practical chemistry, and might profitably be given to students as an easy and pleasant task to render them familiar with oi'dinary weighings and calculation. (a,.) Specific Gravity. — The specific gravity is first taken. This may be done with fciir accuracy by an hydrometer, and still more correctly by a Westphall's balance, or by a specific-gravity bottle. Bottles holding exactly 50 gi-ms. of water at 15° may be pur- chased. The bottle is filled with the milk, first bi'ought to the required temperature, by either cooling or heating, as the case may be, and weighed, and the specific gi-avity obtained by multiplying by -02 ; or if a bottle is used which contains no simple multiple of 100 grms. of water, the ordinary equation may be used,— Weight of water : 1 000 : : weight of milk : specific gravity. Yieth has constructed a useful table (see Table Xllla.) to avoid corrections for temperature. (6.) Total Solids. — The specific gravity having been obtained, exactly 10 cc. are transferred by means of a pipette to a plati- num dish, and submitted to the action of a water-bath, until the contents cease to lose weight; this usually takes from two and a half to three hours. It may be proved by direct experi- ment, that the results from the use of platinum are fiir more constant and more speedily obtained than those obtained by the use of porcelain or glass evaporating dishes. The author has invariably found that porcelain gave a higher result than plati- num, or, in other words, porcelain is more favoural)le to the milkman. When the residue is perfectly dry, it is at once weighed, and the results expressed in percentage by weight. The weight of the 10 cc. is known from the specific gravity already taken. Thus, supposing a milk of 1-032 specific gravity to give a total residue from 10 cc. of l-4'23 grm. : since the specific gravity has shown that 100 cc. of the milk weighed 103-2 grms., it follows that 10 cc. must weisjh 10-32 irrms. TABLE (XIII*) for Correcting the Specific Gravity of Milk according to Temperature. Deosees OP DEGREES OF THERMOMETER {Fahrenheil). 45 190 19-9 32-2 33 46 19-0 20 210 22-0 22-9 23-9 24-9 25-9 26-8 21-0 22 24-9 25-9 50 19-2 20-2 21-2 54 19-5 25-2 26-2 29-8 25-3 26-3 27-3 28-3 29-2 30-2 31-2 25-5 26-5 20 f 21-8 24-7 25-7 29-7 30-6 31 C 20-9 21-9 210 22 230 24 25 260 270 28 29 21-2 22-2 23-2 24-2 210 221 22-5 23 5 27-4 28-4 32-3 33-2 30-8 31-7 32-7 33-7 34-7 34 35-0 32-3 33-3 33-6 34 •( 35-0 33-9 31-9 350 35-2 :ic-2 § 140.] EXAMINATION AND ANALYSIS OF MILK. 239 Hence, 10-32 grms. have yielded a i-esidue of 1-423 grras., which is (calculating only to the first decimal place) 13-7 per cent. (c.) Extraction of Milk-Fot. — On treating this dry residue by ether, or petroleum, a further loss will be perceived when the ether is poured off, and the fat-free residue is first dried and then weighed ; and this loss represents approximately the milk-fat. {(I.) Extraction of the Milk-Sugar. — On now exhausting this residue by weak boiling alcohol, filtering the alcoholic fluid, and evaporating to dryness, the milk-sugar, with mineral matters dissolved out by the alcohol, is obtained. This evaporation is best effected in a platinum dish. On drying very carefully, weighing, and then burning the sugar away, the ash is left, and must be weighed and subtracted from the original residue of milk-sugar and ash. The amount of sugar is thus obtained with fair accuracy, always being a little too high. (e.) Albuminoids and Ash. — Lastly, the casein, albumen, and insoluble ash left from these operations may be carefully dried and weighed, and then burnt to an ash. This ash, subtracted from the total weight, gives the percentage of albviminoids, while the ash from tliis operation, added to the amount of ash from the sugar residue, gives the " total mineral constituents." Thus, in this way, it is quite possible, with care, to make a fairly satisfactory general analysis of milk by using only 10 cc. \iv practice, however, it is found far more convenient and accurate to take two or three separate portions for the analysis : for ex- ample, 10 cc. for the determination of " total solids " and ash ; and 25 cc. for the milk-fat. The foregoing is a brief sketch of the principles of the proxi- mate analysis of a sample of milk. It will be seen that by the simple use of heat, and the application of solvents, such as alcohol and ether, milk is divided into water, milk-fat, milk- sugar, albuminoids, and mineral matters. It is necessary now to consider more in detail the various processes which have been proposed for effecting, in the most accurate and expeditious manner, this division. B. Various Methods Proposed for Extracting the Milk-Fat. Solvents for Fat. — The solvents for milk-fat are petroleum, ether, or bisulphide of carbon. The latter has some advantages when acid milks are analysed, since it has no solvent action on the lactic acid. In quite fresh milks, however, the quantity of lactic acid is so small, that ether may be used. 25 to 50 cc. of milk are evaporated in a flat dish with constant stirring and breaking-up of the caseous films by a glass rod, until the whole 270 FOODS : THEIR COMPOSITIO:^ AND ANALYSIS. [§ 140. is reduced to a rather coarse granular powder. This powder may- be transferred to any simple apparatus in use for the exhaustion of substances by volatile solvents; as, for example, Soxhlet's, described at page 67, — the results are a little low. (a.) Addition of Sand, or SuljjJiate of Lime, or other Poivder. — Numerous experiments by Englisl) analysts on the extraction of milk-fat by solvents acting on dried milk, have fully proved that the greatest accuracy is only obtained by subdividing minutely the dry milk solids; this may be done either by mixing gypsum, sand or ])owdered pumice stone ; the amount of indifferent material used must be at least equal in weight to the milk taken, and it is even advantageous to double the quantity, for instance — 20 grms. of gypsum to 10 grms. of milk. The material is stirred into the milk, dried, and thoroughly well pulverised, the powder being then transferred to a Soxhlet's apparatus, and exhausted by solvents in the usual way. (6.) Adams^ Method of Extracting Milk-Fat. — The Society of Analysts appointed in 1884 a committee to investigate the best methods of milk analysis. The members of the committee analysed by different methods a large number of samples. The results were of considerable interest, especially with regard to extraction of the milk-fat.* The committee clearly established, that milk evaporated to dryness in the ordinary way retains with tenacity a small quantity of fat, and therefore the Wanklyn method is always a little low. The results of the Wanklyn method are better with milks rich in fat than with poor or skimmed milk ; in the latter, the longer the total solids are dried, the less are they amenable to the influence of ether. By finely dividiug the milk, as for example by mixing it into a paste with plaster of Paris, drying, and then exhausting in a Soxhlet, much higher results were obtained. " Milks so highly skimmed that they yielded nothing practically by Wanklyn's method, give upwards of '6 per cent, by the plaster of Paris method ;" but the best results of all were obtained by an ingenious process invented by Dr. Adams — this consists in spreading a measured or weighed quantity of the milk by capillary attraction over a small coil of blotting- or filter-paper, which paper may have to be specially prepared by exhausting it with ether. The coil after having been weighed is charged with milk, and dried at 100° for "total solids," it is then transferred to a Soxhlet. A convenient coil has been suggested by Messrs. Allen k Chattaway f made as follows : — A strip of pai^er, * Analyst, Jan., 1886. t Analyst, April, 1886. §]40.] EXAMINATION AND ANALYSIS OF MILK. 271 21 ins. X 2-5 ins., is threaded down either side by a piece of string, at the end a loop is left and knotted ; the slip of paper is then rolled on a glass rod commencing at the loop end of the slip. When the coil has been rolled as far as the last holes in the paper, the two ends of the strings are tied in a knot as close to the paper as possible. A hole is then made through the centre of the paper immediately under the knot, and of sufficient size to allow the knot to go through. The winding of the remain- ing 3 ins. of the paper is then completed, and a plaited cap of filtering paper placed over the bottom of the coiL* A hole is made through this cap, through which is passed the ends of the string, and the cap and coil are finally secured by tying the two ends of the string round the cap. Special paper cut into strips can now be bought. 5 cc. of the milk are very gradually dropped on to the coil, distributing it as evenly as possible ; it is then dried and exhausted by ether in a Soxhlet. Sour milks, which Dr. Adams has proposed to treat with ammonia so as to liquefy them, admit of analysis by means of this coil without such addition. In the latter case a quantity can be weighed out in a tared dish, and pouring it on to the coil, the dish may be rinsed out by the least quantity of water possible; the washing is of course added to the coil. The general result of the coil process is that it gives results about -5 per cent, higher than the Wanklyn method, and -2 per cent, higher than the plaster of Paris method. To meet tlie obvious objection that the coil method might perchance yield something to ether other than fat, careful analyses were made by the committee of the ether-extracts, both from the plaster of Paris method, and from the coil method ; but in both, fat, and fat only, was found to be extracted. (c.) Estimation of Milk-Fat hy Centrifugal Machines. — One of the earliest, " the lactocrite," was patented by Carl G. P. de Laval (1885, No. 8610). The milk-fat is separated from the milk by strongly acidifying and rotating the milk in a special tube in a centrifugal machine. The tube is thus described in the specification — the test vessel consists of a cylindrical silvered metallic box («), into which a plug,, likewise silvered, (h) fits accurately. This plug is turned and bored hollow from the bottom, tapering towards the top into a fine hole communicating with a narrow tube (c) fixed to the plug, and either graduated or not graduated. This glass tube is fixed upon the plug by the nut {d) received into the socket or * Or the paper may be simply coiled, and then completed by a plaited filter cap. 272 FOODS : their compositiox and analysis. [§ 140. liolcler (e) secured to the plug and surrounding the tube, and provided with opposite openings so that the degrees on the tube may be read. The nut is provided with a hole [g), so that the plug and the tube constitute conjointly a vessel or channel open at both ends (fig. 33). Fig. 3.S. The milk is prepared as follows: — An eqiial volume of milk and u mixture of acetic and sulphuric acids (20 vol. of glacial acetic and 1 vol. cone, sulphuric acid) are placed in a test tube provided with a cork, through which a tine glass tube (to allow for expansion) is carried, and the tube with its contents is then heated in a water-bath for live or six minutes, the special tube is then charged and transferred to a recess on a metallic disc {x), previously brought by hot water to the tempei'ature of 50^^. The disc is usually fitted with twelve such recesses, so that 12 samples of milk may be operated upon at one and the same time. The disc is rotated at a speed of GOOO revolutions per minute; after from three to five minutes the milk-fat will have risen in a well-defined layer; the reading is done at once while the tubes are warm. The divisions are in tenths, thus 33 '5 is equal to 3*3o per cent, of fat in the milk. 140.] EXAMINATION AND ANALYSIS OF MILK. 273 The lactocrite is not much used now, its place being taken by various centrifugal machines, the most favoured being the Leffman and Beam and the Lister-Babcock. The disc in these does not require to be kept specially warm, the casein is brought into solution by strong hydrochloric and sulphuric acids, and the separation of the fat aided by a certain proportion of amyl alcohol. There are printed directions supplied by the makers of each instrument. TABLE XI IB. Comparison op Centrifugal Determinations op Milk-Fat witu Adams' Processes. i 's- 2; o 1 m m ."2 Fat. Solids koi Fat. Ash. 2 i < -A, III .■§■ 1 2 ■a < Ill >> 1. 1033-0 12-90 3-40 3-50 3-44 3-42 9-50 9-40 9-46 9-48 -80 2. 10310 1412 4-55 4-05 4-69 4-62 9-57 9-47 9-43 9-50 •80 3. 1032 1312 3-60 3-57 3-62 9 52 9-55 9-50 -77 4. 1030-5 13 07 3-90 3-99 3-89 9-17 9 08 9-18 •74 5. 1032 12-98 3-80 3-76 3-51 9-18 9-22 9-47 •76 6. 1031-5 14-27 4-90 4-84 4-63 9-37 9 43 9-64 -76 7. 1032-5 13-84 4-20 4-26 4-08 9-64 9-58 9-76 -78 8. 1030-5 13 00 3-70 3-69 3-84 9-30 9-31 9-16 -76 9. 1031-0 13-51 4 05 4-09 4-14 9-46 9-42 9-37 -80 10. 1034-0 11-76 2-10 2 07 212 9-66 9-69 9-64 -76 11. 10.35-0 9-99 •45 •502 •518 9-54 9-488 9-472 •86 19 274 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 140. (d.) Soxhlet's Process. — F. Soxlilet* is tlie author of a very ingenious process of milk analysis. This pi'ocess is based on the fact, that if a measured quantity of milk, alkalised by caustic potash, be shaken up with ether, the ether fully extracts the milk-fat, and, on standing, collects in a clear layer. A small, quite constant, proportion of ether remains in solution in the milk, without retaining any of the fat, and without afiecting the result. The amount of the fat dissolved in the ether may be determined by the specific gravity of the ether ; the higher the specific gravity the greater the proportion of milk-fat. The details of Soxhlet's method are as follows : The apparatus figured (see fig. 34) is used. C is ameasuring flask of 300 cc. capacity, provided with a doubly perforated cork, and connected, on the one hand, with the caoutchovic elasticbulbs figured, which are furnished with suitable valves, and, on the other, with the tube D, which is provided with a water- jacket. This tube carries an areometer, E, which bears a delicate scale of from -766 to -743, and it has also a thermometer, divided into thirds of a degree. 200 cc. of milk are measured by means of a pipette, and run into the flask; 10 cc. of potash and 60 cc. of ether, which has been saturated with water at from 16^-5 to 18°-5, are then added, and the whole shaken up in the properly closed flask for a quarter of an hour. The fluid is then allowed to repose until the ether rises in a clear layer to the surface. By gently working the caoutchouc bulbs, a sufficient quantity of the fat-laden ether may now be blown up into the tube D, to float the areometer. Fig. 34. * Zeitschri/t der Landmrthsch.-Ver. Bayern. ISSO. § 141.] EXAMINATION AND ANALYSIS OP MILK. 275 (It should be mentioned that the tube going through the perfo- rated cork of C, is so arranged as to dip well into the ether, but does not touch the surface of the milk.) The areometer must float freely, and for this purpose there are little prominences on the inside of D, about the middle, for the purpose of keeping the instrument from adhering to the side. When sufficient ether has thus been blown up, it is retained by a pinch-cock, and a cork is fitted into the tube D, to prevent loss by evapoi-ation. The water in the jacket must be of any temperature between 16°-5 and 18° -5. The specific gravity once in this way obtained, and the temperature of the areometer noted, the content in fat of the milk is determined by the annexed table (p. 276). If the temperature is exactly 17° "5, then no correction is required; but if it is above that teuiperature, for every degree, a degree must be added to the gravity ; if below, for every degree, a degree must be subtracted. Thus, supposing the areo- meter marks 58-9 at 16°-8, then, as the difi"erence between 16°-8 and 17°-5 is -7, subtract u from 58-9, equalling 58*2 ; or if the observation be 47 G at 18°-4, then the corrected value is 48*5 {i.e., specific gravity -74850). Afterwards the ether is allowed to run back into the flask,, and D is washed out with a little ether. The author thinks that the apparatus might be improved by adapting the prin- ciples used in the tube figured at p. 69, and floating the ether layer by means of mercury. (e.) The Werner- Self inidt method of fat estimation'^ gives accurate results. A test tube of 50 cc. capacity is calibrated in cubic centimetres and tenths; into it are placed 10 cc. of milk, or a weighed quantity (say 5 grms.) of cream, 10 cc. of strong hydro- chloric acid added, and the whole boiled, with shaking until the liquid forms a dark brown solution. The tube, with its contents, is rapidly cooled by placing it in cold water ; a known volume of ether is added by filling it up to one of the divisions, some 25 to 30 cc. of ether being necessary. The contents are now vigorously shaken for half a minute, and the tube is allowed to stand for five minutes. At the end of that time the layer of ether is carefully read, and an aliquot part pipetted off", evapo- rated to dryness, and weighed. 0. Various other Methods of Milk Analysis. § 141. There are other methods of analysing milk : — (1.) Drying in a Vacuum. — The author finds that by putting 5 cc. of milk in a small strong flask, and connecting this liask with a second containing strong sulphuric acid, and, lastly, attaching the two flasks to a Lane-Fox mercury pump, the milk solids may be boiled, and the water expelled otf at a temperature of * Zeitschrift f. analytische Chan., vol. xxvii. ; see also A. W. Stokes, in Analyst, Feb.,"l889. 276 . TABLE XIV.— Correction of Sped fie Gravity of the Ethereal Solution of Fat (pp.' 274, 275). Specific Fat. Per Specific Fat. Per Specific Fat. Per Specific Fat. Per Specific Fat. Per gravity. cent. gravity. Cent. gravity. cent. gravity. cent gravity. cent. 43 2-07 47-7 2-61 52-3 3-16 56-9 3-74 615 4-39 43-1 2-08 47-8 2-62 52-4 3-17 57 3-75 61-6 4-40 43-2 2 09 47-9 2-63 52-5 3-18 57-1 3-76 61-7 4-42 43-3 210 48 2-64 52-6 3-20 57-2 3-78 61-8 4-44 43 4 2-11 48-1 2-66 52-7 3-21 57-3 3-80 61-9 4-46 43-5 212 48-2 2 67 52-8 3-22 57-4 3-81 62 4-47 43-6 2-13 48-3 2-68 52-9 3-23 57-5 3-82 62-1 4-48 43-7 2-14 48-4 2-70 53 3-25 57-6 3-84 62-2 4-50 43-8 2-16 48-5 2-71 531 3-26 57-7 3-85 62-3 4-52 43-9 2-17 48-6 2-72 53-2 3-27 57-8 3-87 62-4 4-53 44 2-18 48-7 2-73 53-3 3-28 57-9 3-88 62-5 4-55 44-1 2-19 48-8 2-74 53-4 3 29 58 3-90 62-6 4-56 44-2 2-20 48-9 2-75 53-5 3-30 58-1 3-91 62-7 4-58 44-3 2-22 49 2-76 53-6 3-31 58-2 3-92 62-8 4-59 44.4 2-23 49 1 2-77 53-7 3-33 58-3 3-93 62-9 4.61 44-5 2-24 49-2 2-78 53-8 3-34 58-4 3-95 63 4-63 44-6 2-25 49-3 2-79 53-9 3-35 58-5 3-96 63-1 4-64 44-7 2-26 49-4 2-80 54 3-37 58-6 3-98 63 2 4-66 44-8 2-27 49-5 2-81 541 3-38 58-7 3-99 63-3 4-67 44:9 2-28 49 6 2-83 54-2 3-39 58-8 4-01 63 4 4-69 45 2-30 49-7 2-84 54-3 3-40 58-9 4 02 63-5 4-70 45 1 2-31 49-8 2 86 54-4 3-41 59 4-03 63-6 4-71 45-2 2-32 49-9 2-87 54-5 3-43 59-1 4-04 63-7 4-73 45-3 2-33 50 2-88 54-6 3-45 59-2 4 06 63-8 4-75 45-4 2-34 50 1 2-90 54-7 3-46 59-3 4 07 63-9 4-77 45-5 2-35 50-2 2-91 54-8 3-47 59-4 4 09 64 4-79 45-6 2-36 50-3 2-92 54-9 3-48 59-5 411 64-1 4-80 45-7 2-37 50-4 2-93 55 3-49 59-6 4-12 64-2 4-82 45-8 2-38 50-5 2-94 55-1 3-51 59-7 4-14 64-3 4-84 45-9 2-39 50-6 2-96 55-2 3-52 59-8 415 64-4 4-85 46 2-40 50-7 2-97 55-3 3-53 59-9 4-16 64-5 4-87 46-1 2-42 50-8 2-98 55-4 3-55 60 4-18 64-6 4-88 46-2 2-43 50-9 2-99 55-5 3-56 60-1 4-19 64-7 4-90 46-3 2-44 51 3 00 55-6 3-57 60-2 4-20 64-8 4-92 46-4 2-45 51-1 301 55-7 3-59 60-3 4-21 64-9 4-93 46-5 2-46 51-2 3 03 55-8 3-60 60-4 4-23 65 4-95 46-6 2-47 51-3 3-04 55-9 3-61 60-5 4-24 65 1 4-97 46-7 2-49 51-4 3 05 56 3-63 60-6 4-26 65-2 4-98 46-8 2-50 515 3 06 56-1 3-64 60-7 4-27 65-3 5-00 46-9 2-51 51-6 3-08 56-2 3-65 60-8 4-29 65-4 5 02 47 2-52 51-7 3 09 56 3 3-67 60-9 4-30 65-5 5 04 47-1 2-54 51-8 3-10 56-4 3-68 61 4-32 65-6 5 05 47-2 2-55 51-9 311 56-5 3-69 61-1 4-33 65-7 5 07 47-3 2-56 52 3-12 56-6 3-71 61-2 4-35 65-8 5 09 47-4 2-57 52-1 3-14 56-7 3-72 61-3 4-36 65-9 5-11 47-5 2-58 52-2 315 56-8 3-73 61-4 4-37 66 512 47-6 2-60 N.B. — The numbers in the specific gravity column correspond to those on the areometer scale, the number 7 being omitted ou account of the narrowness of the stem of the instrument ; thus 46 really means '7460, and so on with the rest. § 141.] EXAMINATION AND ANALYSIS OF MILK. 277 about 40°. The solids are perfectly dry in about one hour and a quarter. It was noticed in some test-experiments that in all cases the dry solids obtained in this way were about -5 per cent, higher than the weight of dry solids in the water-bath — a proof that in all cases drying in the ordinary way entails loss. After the dry solids have been deprived of fat by ether or petroleum, the albumen and sugar may be dissolved out by cold water to which acetic acid has been added. On precipitation of the albumen by boiling, the latter may be collected on a weighed filter, dried, and weighed. On evaporation of the filtrate, the soluble ash, the milk-sugar, and the bodies described at p. 248 are left ; on ignition, the difterence of weight before ancl after gives vei-y nearly the amount of sugar. Lastly, the casein and insoluble ash are obtained from the portion of original milk insoluble in cold water. (2.) Direct Determination of the Water. — In most analyses the water is inferred from the loss ; it may, however, occasionally be necessary to estimate it directly. This can be readily done as follows: — 5 cc. of milk are placed in a small flask; a piece of tubing is sealed at one end, and graduated into cc.'s ; it will only be necessaiy to mark it at 4-5 and at 5-5, graduating it, between these numbers, into lOths. The tube is now bent twice at right angles, and connected by a caoutchouc cork to the flask ; another narrow tube goes to the mercury pump ; the flask is now exhausted of air ; and by applying a flame to the connecting tube, the tube is drawn out and sealed. By now plunging the limb of the graduated sealed tube into ice and salt, and gently warming the flask, the milk boils, the water is all condensed in the limb, and the amount can be seen by simple inspection. A somewhat similar method of analysing milk, by a special apparatus, has been patented in Germany, by J. Petri and R. Muencke.* (3.) Absorption of Water hy Dehydrating Agents. — Another method for the general analysis of milk consists in adding an excess of anhydrous gypsum to the milk, which very rapidly extracts the water ; the powder can then be exhausted by ether, and the fat estimated, while, similarly, the sugar may be dis- solved out by alcohol. This method for the purpose of estimation of fat and sugar is most decidedly rapid and convenient. (4.) Rittliausert s Copper Process. — Ritthausent dilutes milk to twenty times its volume, and adds a solution of copper acetate, * Deutsche Patentschrift, No. 7477. + " Neue Methods zur Analyse der Milch," and " Ueber ein vom Milch- Zucker verschiedenes Kohlenhydrate in der Kuhmilch," von H. Ritthausen, Journal fur Prak. Chemie, 15, p. 329, 1877. 278 FOODS : their composition and analysis. [§ 141. 10 cc. = -0095 CuO [or copper sulphate, 10 cc. = -2 CuO]. A sufficient solution of KHO or NaHO must also be added, to prevent the separation of basic salt. On filtration, a filtrate separates containing no copper, and may be used for the esti- mation of sugar by Fehling's solution; the precipitate is dried and exhausted of fat, and ultimately ignited. This method has not come into general use, but in certain cases it has advantages. (5.) Miiller's Process. — Alexander Miiller* is the author of a method of milk analysis which is ingenious, but it is difficult to see in what way it is superior to other processes more used, either in precision or accuracy. He uses small flasks, the weight of which must not exceed 25 grms., which will hold when full about 60 to 65 cc. In one of these small flasks 6 cc. of the milk are put and weighed ; exactly 50 cc. of a mixture of 1 volume absolute alcohol and 3 volumes ether are now added, with special precautious to prevent loss by evaporation. The solvent is allowed to act for twenty-four hours. The precipitate by this time has separated completely, and it is possible to isolate the alcohol-ether contents without filti'ation. The next process is to obtain exactly 50 cc. of the liquid without loss and without the precipitate. This is eftected by connecting the digestion flask by means of a bent tube with a flask which is marked for 50 cc. on the neck. This last flask has a double bored cork, one carrying the connecting tube, the other a tube for the purposes of suction ; and the ether-alcohol is carefully sucked over to the mark, gently inclining the digestion flask, &c. Lest the incoming air should carry away any of the alcohol-ether, it (the air) is made to pass first through a few cc. of ether in a third flask. The 50 cc. are evaporated to dryness, and contain fat, with a little milk-sugar and a little common salt. For technical purposes, a correction constant is made for these impurities of about one-quarter per cent. ; while for scientific purposes, the fat is taken up again by means of petroleum ether. The residue in the digestion fiask contains nearly all the milk-sugar, the mineral constituents, and the albuminoids; it is dried and weighed in the usual way. The fiit from the one flask and the solids from the other, when added together, equal the total solids. Miiller calculates the casein by determining the nitrogen, and obtains the ash by acidulating 6 cc. with a little nitric acid, heating slowly with a heat that is gradually increased from gentle warmth up to redness.f * Zeitschrift fur Analytische Chemie, xx, 189. + It is necessary to observe that in extracting the fatty constituents by § 141.] EXAMINATION AND ANALYSIS OP MILK. 279 (6.) Clausnizer and A. Mayer s Process. — An indirect method of estimating milk-fat has been proposed by F. Clausnizer and A. Mayer.* They assume that every percentage of "milk solids not fat" raises the specific gravity -00375, whilst every 1 per cent, of fat lowers the specific gravity -0010; by therefore first making an accurate determination of the specific gravity, and then determining either fat or total solids, the equation can be easily solved. ^ = ^^*'.^ •. «. -00375 + 1 - s. s = specific gravity. x = .„..,„g T t = total solids. •00475 Eichmond % has recently proposed the following formula : — t =1 +|a: + 0-14. 4 5 A milk scale calculated on a similar formula, known as Rich- mond's milk scale, is a useful instrument, saving calculation, and can be bought for a few shillings. Summary of the Processes for the Technical Analysis of Milk already described. What particular procedure an analyst may select to obtain a knowledge of the general composition of milk, will mainly depend upon the purposes for which the analysis is required. Where a large number of milks are analysed daily the most convenient technical method, whether for sorting milks into adulterated and genuine, or for the control of a milk supply, is the following : — ^The specific gravity is taken by a hydrometer ; the samples are then prepared for "whirling" by the addition, to measured quantities, of the proper amounts of hydrochloric and sulphuric acids and amyl alcohol, and "whirled;" the quantity of fat having been thus obtained, the total solids are calculated out by Richmond's formula or by the aid of the milk scale. The milk-sugar is determined by polarisation and by this ether process, when 50 cc. of ether-alcohol are added to 6 cc. of milk, the c. volume is not 56 cc, but 54-3 cc. ; therefore, the weighed fat must be multiplied by ^^ in order to obtain the quantity of matters dissolved from 6 cc. * F. Clausnizer and Adolf Mayer. Forschungen auf dem Gebiete der Viehhaltung in ihrer Erzeugnisse, 1879, 265. t Another formula in use is that suggested by Fleischmann, the symbols are as above. T . 1 o , oftR.^100s - 100 I. t — \-2x + 2-665 : . II. X = -SZU X Analyst, xx., 57. 280 foods: their composition and analysis. [§ 142. copper as described on pp. 282, 283. These determinations will certainly sufficiently denote to the analyst whether he has a pure or an adulterated milk to deal with ; if the latter, he will, of course, analyse the milk very thoroughly, but if in this pre- liminary testing the milk is shown to be good, he may consider it a waste of time to proceed farther. IIT, Special Details as to the More Exhaustive and Scientific Analysis op Milk. § 142. (1.) Analysis of the Milk-Fat and Examination of the Ethereal Extract of Milk. — The milk-fat is seldom analysed, save when met with in commerce as butter. Yet, in the complete examination of milks suspected to be abnormal, it is of some importance to examine the ethereal extract as thoroughly as possible. The ethereal extract should be tested for nitrogenous substances, for phosphorus compounds, for cholesterins, and finally saponified, and the volatile acids at least estimatfid.* Nitrogenous matters may be assumed to be absent if, on boiling the fat with a strong alkaline lye (proved to be ammonia-free) and distilling, an alkaline distillate is not obtained. Phosphorus may be tested for by burning up the fat intimately mixed with a mixture of carbonate and nitrate of soda, dissolving the ash, acidifying with nitric acid, and testing with molybdate of ammonia. Any jirecipitate which occurs is allowed to stand for twelve hours, and is then filtered off, dissolved in dilute ammonia, and precipitated by the ordinary magnesia mixture. Thus, the experiment may be made quanti- tative as well as qualitative, for the ammonio-magnesia phosphate may be dried, ignited, and weighed. The further analysis of the milk-fat is described in the article on "Butter." Cholesterin (C^gH^^O + HgO). — Cholesterin has been found in gall stones, eggs, the spermatic fluid, the nervous tissue, milk- fat, and in nearly all animal oils. It is soluble in hot alcohol, ether, carbon disulphide, and chloroform. It polarises to the left, and has a melting point of 145° when anhydrous. It crystallises in white plates or monoclinic tablets containing one atom of water of crystallisation, but it separates from chloroform water free. The crystals treated with a mixture of five parts of sulphuric acid and one of water are coloured first carmine red and then violet (Moleschott's test). A solution of the crystals in chloro- * Very rarely, it would seem, that milk-fat contcains a free fatty acid ; in a case related by C. Arnold, Arc/i. PItarm. [S], xx., 291-293, the etlier extract contained '8 per cent, of free fatty acid. § 143.] EXAMINATION AND ANALYSIS OP MILK. 281 form on the addition of an equal volume of sulphuric acid is coloured blood red, the colour changing into violet red, and lastly into purple (Salkowsky's test). If the crystals are dis- solved in hot acetic anhydride and, when cool, a few drops of concentrated sulphuric acid added, an enduring blue colour results. Phyto-cholesterin (CogH^^O + B..,0) is similar in properties to cholesterin, but has a lower melting point (132° to 133°); phyto-cholesterin occurs in vegetable oils, not in animal ; * a small quantity of phyto-cholesterin has, however, been found in milk-fat. Cholesterin is detected in milk-fat or other fat as follows : — 10 grms. of the pure fat are saponified by an equal weight of KHO dissolved in 10 cc. of water and 10 cc. of alcohol ; after saponification the solution is diluted with 600 to 700 cc. of water and shaken up in a large separating cylinder with 500 cc. of ether. The ether is separated and recovered by distillation. The residue, which always contains some fat, is again saponified on the same lines as before and the solution of soap diluted and shaken up again with ether. The ether is separated and shaken up several times with distilled water, and finally separated and recovered by distillation. The residue is dissolved in a little hot alcohol and concentrated down to 2 cc. On standing, the solution will deposit crystals, the melting point of which may be taken and tests applied. § 143. (2.) Detection and Estimation of the Carbo-hydrates of Milk. — Tlie first step is to prepare a clear whey. An excellent paper on this subject was published in 1887 by Dr. W. Wiley, the chief chemist for the U.S. Department of Agriculture.! He used the following reagents for the precipitation of the albuminoids : — (1.) Saturated solution of basic lead acetate. (2.) Nitric acid solution of mercuric nitrate diluted with an equal volume of water. (3.) Acetic acid, specific gravity r040. (4.) Nitric acid, specific gravity ri97. (5.) Sulphuric acid, specific gravity 1-255. (6.) Saturated solution of sodium chloride. (7.) Saturated solution of magnesium sulphate. * Salkowsky has suggested the separation of cholesterin from oils as a test of genuineness. Kape and cotton seeds, for example, always contain phyto-cholesterin ; cod-liver oil, cholesterin ; he found pure cod-liver oil ta contain a cholesterin which melted at 146°, while a sample of oil adulterated with 20 per cent, of linseed oil gave a cholesterin which melted at from 139" to 140°. Hence, the process may be valuable in detecting mixtures of animal and vegetable oils. Ztit. f. anal. Chemie, 1887, Bd. 2G, S. 567. t Foods and Food Adulteration, United States Department of Agricul- ture, Bulletin No. 13. 282 foods: their compositiok and analysis. [§143. (8.) Solution of mercuric iodide in acetic acid, a litre of the solution being made •with 161 grms. HgCl2, 395 grms. KI, and 228 cc. of acetic acid. Dr. Wiley also used a number of other precipitating agencies, but finally gave the preference to two solutions, viz., to the mercuric nitrate, or to the mercuric iodide solution. These solutions he used as follows : — He added 1 cc. of the mercuric nitrate, or 30 cc. of the mercuric iodide solution, to certain quantities of milk, varying these quantities according to specific gravity. For a milk of specific gravity 1'026 he took 60-5 cc. of the milk ; if of specific gravity 1*030, 60 cc. of milk ; and if of specific gravity 1-034, 59-5 cc. of milk. The precipitated albuminoids he considered to occupy a bulk of 2-4 cc, so that the volume in each case was made up to 102-4 cc. The milk thus prepared was shaken, filtered, and the filtrate polarised. The instrument used by Wiley was a Laurent large model polarimeter. The specific rotatory power of milk-sugar for the sodium light he considered to be 53°. The author has tried both the mercuric nitrate and tha mercuric iodide precipitants, and both give good results ; but the presence of mercury in the filtrates is objectionable, for the filtrate can only be used for the single purpose of polarimetrical observation. For shadow instruments quite as clear and bright a filtrate can be obtained by the use of copper sulphate, and if only the proper quantity is taken the filtrate is of so pale a blue that it in no way interferes with the reading ; besides which, such a liquid can be treated with Fehling, and the amount of copper suboxide reduced ascertained by weighing. The author precipitates as follows : — 25 cc. of milk are diluted to about 50 cc. with distilled water, and then strong acetic acid is added drop by drop until the casein begins to separate. The liquid is then heated to boiling, and while still hot, " whirled " in a glass cylinder by the aid of a centrifugal machine. The casein, coagulated serum, albumen, and fat separate and collect in a more or less firm coagulum at the bottom of the cylinder, and the supernatant fluid is easily filtered ; the precipitate is also transferred to the filter and washed with hot water. Finally, the filtrate is cooled, made up to 100 cc, and submitted to the processes to be detailed. The filtrate is usually, in a normal milk, of a feeble yellow colour, but perfectly bright, and suitable for testing for sub- stances such as salicylic acid, borax, dextrin, or any soluble addition. It is also most suitable for optical methods, and for any estimation of sugar by the reduction of cupric oxide. It § 143.] EXAMINATION AND ANALYSIS OF MILK. 283 will be obvious that such a solution should contain somewhere near 1 per cent, of hydrous milk sugar. The optical power of sugars generally is not the same for strong as for dilute solutions, so that as a 1 per cent, solution is in all these cases taken, it is essential that the instrument be standaixlised with the same solution as is used for standardising the copper process. For example, in the instrument used by the author, 10-28 per cent, milk-sugar in a 400 mm. tube marks 100° on the scale, but a 1 per cent, marks 10°"1 ; similarly, 1 per cent, of cane-sugar, or 1 per cent, of dextrose, all have for such dilute solutions a slightly different value at ordinary temperatures — from 15° up to 18°— than stronger solutions. If cane-sugar is large in amount relatively, the optical supple- mented by the copper process at once reveals it, for we have in this case two sugars, both having an action on polarised light in different proportions, and the one having an action on copper, the other but little action on copper. Thus, a milk which was prepared so as to have practically equal parts of cane- and milk- sugar, containing in 100 cc. 3-48 of cane-sugar and 3-52 of milk-sugar, gave by the saccharometer an indication of 8 per cent, of milk-sugar ; by copper an indication of 3*9 per cent, of milk-sugar, so that, without inversion, it would be evident that in this case a sugar was present that rotated strongly, and only acted on copper slightly; and dextrin having been proved absent by testing the clear liquid with iodine, there would not be much error in diagnosing the presence of cane-sugar. With smaller proportions of cane-sugar to milk-sugar, the differences are not so great, and inversion is necessary. Inver- sion of the whey of pure milks, calculated, not into galactose, but lactose, makes an apparent difierence of from 0'5 to 0'8 per cent. ; thus, a milk which by the optical method showed 4-11 per cent, of milk-sugar, and by copper 4-23 per cent., on inversion (by HCl) showed 3-5 per cent, milk-sugar at 16°. When cane-sugar is suspected, it is best to invert by citric acid or by invertase ; 1 grm. of citric acid is added to 100 cc. of the filtrate, and the whole boiled for twenty minutes ; the cane- sugar is then inverted, the milk-sugar not being affected. If glucose be added to milk, we then have the problem of two sugars, which act unequally to polarised light and which act unequally towards copper. The following table gives an example of a few milks which have been "whirled," the fat determined by the centrifugal machine, the specific gravity taken by a hydrometer, the total solids calculated by the Richmond scale, and the milk-sugar determined both by copper and by the polarimeter : — 284: foods: their composition and analysis. [§ 14o. TABLE XV.— Analyses of Commercial Milks. 1 Specific Gravity. Total Solids. Fat. Milk-sugar. Polarimeter. Cyanide Copper Process. 1,031 1,034 1,032 1,033 ],032 1,034 1,033 1,033 1,031 10-4 12-68 111 12-9 12-3 12-55 11-85 12-78 11-8 2-2 3-5 3 3-8 3-6 3-4 3-0 3-8 3-4 4-08 5-00 4-24 4-24 4-04 4-40 4-24 4-72 4-20 4-31 4-90 4-31 4-26 4-56 4-24 4-23 4-92 4-26 Hence, it is abundantly clear that, should there be an appreciable difference between the presumed percentage of milk-sugar, as determined by optical and copper processes, the carbo-hydrates of the milk are abnormal, and there is a probability of either dextrin or cane-sugar, or glucose, having been added, and it is necessary to examine further. Schmoeger has also found that polariscope determinations of milk-sugar agree very well with the gravimetrical if the serum is prepared as follows : — To 100 cc. of milk, 6 cc. of a 10 per cent, solution of acetic acid are added and allowed to stand for half an hour, and then filtered; to tlie somewhat turbid filtrate, 4 cc. of subacetate of lead solution (specific gravity 1-2) are added, the mixture boiled, cooled, made up to the bulk before boiling, filtered and polarised. Schmoeger found it only necessary to apply a small correction for the alteration of volume caused by the addition of the precipitating solutions ; this correction is made by raising the indication of the polariscope in the propor- tion of 103 to 108. The result is in percentages of milk-sugar in the crystallised state containing 5 per cent, of water. Dextrin may be dismissed in a sentence, for the clear filtrate, when tested with a droplet of iodine solution, gives a character- istic reddish colour (see also p. 130). To identify foreign sugars, such as glucose or eane-sugar, it is best to obtain the osazones by the interaction of phenyl-hydrazine. The osazones may, for the purposes of identification, be divided into two classes — T. Soluble with difiiculty in hot water j II. Fairly soluble in hot water. To the first class belong glucosazone and galactosazone. To the second, lactosazone (and maltosazone). § 143.] EXAMINATION AND ANALYSIS OP MILK. 285 Both classes are almost insoluble in ether, benzol, or chloro- form, and both are soluble in hot acetic acid. Cane-sugar gives noosazone; it has first to be inverted; it then gives an osazone not to be distinguished from glucosazone ; the properties of the osazones likely to be obtained from milk are as follows : — • Lactosazone when slightly impure consists of an aggregation of warty masses; many of the individual gianules are much like starch grains, exhibiting a split hilum with radiating fissures ; and the melting point is from 190° to 195°; when pure, lactosa- zone crystallises in needles, and then has a melting point of about 200°, at which temperature it decomposes and evolves gas. It is soluble in about 90 parts of boiling water ; but not very soluble in cold. Lactosazone shaken up with cold water and allowed to stand for 24 hours only dissolves in the proportion of 20 mgrms. per 100 cc. It is freely soluble in hot absolute alcohol. Lactosazone forms an anhydride, C24HgQ]Sr^O-, ; the anhydride is obtained by the action of very small quantities of mineral acids on milk-sugar, and probably under other conditions. It appears in long silky yellow crystals, almost insoluble in hot ■svater, and melts at the high temperature of 223° to 224° C. without evolution of gas. Galactosazone separates in needle-like crystals, soluble with difiiculty in hot water; it is easier soluble in hot absolute alcohol than glucosazone. The crystals darken at 188° C. ajid melt, with evolution of gas, at 193° to 194° C. Galactosazone is nearly always obtained in small or large quantity from milk whey, whatever process is followed. Glucosazone is never in warty crystals, but, on account of its great insolubility* in hot water, separates early in star groups of radiating needles. Besides its insolubility in hot water, it is not easily dissolved by hot absolute alcohol, but easier by hot diluted alcohol (60 percent.), from which it separates in large crystals ; the melting point, when pure, is 204° to 205°. To obtain the osazones trom milk, 100 cc. of diluted milk whey prepared, as previously described, from 25 cc. of milk, and, therefore, representing approximately 1 grm. of milk-sugar, are neutralised with soda, if acid, and 2 grms. of sodic acetate, and 1-5 grms. of ]»henyl-hydmzine added, and the whole heated in a flask for an hour and a-half ; it is then filtered while hot from any precipitate that has formed. The precipitate may be glucosazone or galactosazone or a mixture; by heating it with hot absolute alcohol, if a mixture of the two, they may be in part separated. The anhydi-ide may also come down in a hot * According to E. Laves, 100 parts of boiling water dissolve 10 mgrms. of glucosazone. Archivder Pharm., ccxxxi., 366. 286 FOODS : THEIR COMPOSITIOX AND ANALYSIS. [§ 143. solution ; by fractional crystallisation, by the differences in solubility in hot water and in hot absolute alcohol, and finally, getting the melting points of the products and observing the shape of the crystals, the osazones are identified. A. Maquenne* has proposed identifying the sugars by their behaviour with phenyl-hydrazine under equal conditions ; if to a 1 per cent, of sugar is added 5 cc. of a solution containing in 160 cc. 40 grms. phenyl-hydrazine and 40 grms. acetic acid, and the sugar solution be heated for an hour at 100°, and allowed to cool ; and if, further, the precipitate be collected on a filter washed with 100 cc. of water and dried at 110°, the following are the weights of the osazones and the behaviour of the solutions. Weight cf Osazone. Time of Tuibidity. Lactose, Galactose, . Glucose, Laevulose, . Maltose, 0-11 0-23 0-32 0-70 0-11 Only on cooling. Precipitate in 30 minutes. ,, 5 ,, Only on cooling. C. E. Guignet f has proposed to distinguish the sugars by an ammoniacal cop23er salt. Finely-powdered copper sulphate is added little by little to strong ammonia, and the liquid ulti- mately boiled for a few minutes. The compound crystallises out on cooling. A solution of this salt precipitates neither cane- nor milk-sugar. The same statement is made as regards laevulose and invert sugar ; dextrose and galactose are after a few minutes precipi- tated ; mannite and sorbin give blue precipitates not changed by boiling. Theodor Selivanoff J uses resorcin as a test; warming aqueous solutions of cane-sugar, laevulose, and ralBnose with resorcin and concentrated hydrochloric acid gives a red colour ; lactose, galactose, maltose, dextrose, and inosite give no colour. § 144. (3.) 7 he Ash. — The ash is estimated in a quantity of milk, which should not exceed 25 cc. It will be found that larger quantities do not at all conduce to accuracy, as the large amount of carbon, with even great care, develops too much heat, and the phosphates are liable to fuse enclosing little particles of charcoal extremely difficult to burn. With small quantities, however, the milk raj^idly burns to an almost white ash. It may be further analysed on the principles laid down at p. 119. * Comptes Rendus, cxii., 799. t Comptes Rendxis, cix., 528. J Btr. d. deutach. chem. Gesellsch., xx., 181. § 145.] EXAMINATION AND ANALYSIS OF MILK. 287 § 145. (4.) Estimation of Albumen. — This may be done when the milk solids are dried in a vacuum, as described at p. 'lib, then the albumen may be dissolved out by tlie aid of cold water, acidulated by acetic acid. If this process is not adopted, the following may be used : — 100 cc. of milk are divided into three equal portions. One of these portions is diluted to about four times its volume, and acidified with dilute acetic acid until tho casein coagulates in a fiocculent condition; a current of carbon dioxide is now passed through, and the precipitate allowed to subside. The whey is then carefully syphoned off on to the second portion of the milk: more acid, if necessary, is added; the same operation repeated; and this second whey similarly added to the third portion of milk. Finally, the wliole of the casein is collected on a filter, and washed. The result of the process is, that the albumen from 100 cc. of milk is held in solution in about 250 to 300 cc. This solution is now raised to the boiling point, and gently boiled for a few minutes. The whole of the albumen falls down, and is easily collected on a previously dried and weighed filter. This easy separation of the casein and albumen by acetic acid and carbon dioxide only applies to the milk of the cud-chewers; with jjuman milk, the milk of the horse or of the ass, the process gives no good result: when treated in the same way, it is true that the casein appears to coagulate, but is in a state of such fine division tliat nearly all of it remains suspended in the liquid, and filtration through paper becomes impossible. From all these a clear filtrate may, however, be obtained by filtration under pressure through one of the Pasteur or Berkfeldt candles, or the globular- shaped vessels formed of kaolin and quartz, as described by W. PukalL* The globular shape is somewhat better than the cylindrical. In all these cases the filtration is upwards ; by means of an angle tube and caoutchouc-pressure tubing, it is attached to a mercury or water pump. The milk is diluted, acidified with acetic acid, and saturated with carbon dioxide as before ; the filter is then totally immersed in the dilute acid milk, which for this purpose is placed in a tall beaker ; a good vacuum is maintained, and ultimately the whole of the whey passes through, and the casein may be washed two or three times with water. Besides this very convenient method, there are many substances which will carry down the precipitated casein mechanically; thus, a solution of phosi)hate of lime in acetic acid may be added to the acid milk; when, on cautiously neutralising with ammonia, the earthy precipitate clears tho liquid by its * Ber. chemlnch. GeselL, xxvi. , 1159-1172. 288 foods: their composition and analysis. [§§ 1-iG, 147. mechanical action. (See also Sebelien's method of estimating casein and albumen, p. 31 1.)* § 146. (5.) Isolation of Peptone-like Body (Galactin). — As in the previous operation, 100 cc. of milk are greatly dihited, the casein •coagulated, and the whey separated by subsidence. Tliis whey is used for the precipitation of a second portion, and the same ■whey from the second pr rtion for the precipitation of a third, and so on. By this means it is possible to obtain the whey from a litre of milk in a form not too dilute. The albumen may be separated from time to time in the diflerent fractions, or in one concluding operation; lastly, the casein must be collected and well boiled, and the liquid separated from it by filtration and strong pressure. A solution of nitrate of mercury is now added, the precipitate separated and decomposed by hydrogen sulphide, and the well-boiled filtrate precipitated by lead acetate. The amount of the peptone may be estimated from the weight of the lead oxide left on ignition of the lead compound, or the filtrate from the mercury sulphide may be treated by Kjeldahl's process, and the amount of peptone calculated from the nitrogen. (6.) Estimation of the Total Nitrogen in Milk. — Kjeldahl's f method of determining nitrogen is applicable to milk : — 5 cc. of the milk are digested with 3 cc. of strong sulphuric acid at 100° for about an hour, then crystals of powdei-ed potassic perman- ganate are added cautiously, and the heating continued until the mixture assumes a green colour. On cooling, the liquid is made up with water to a litre, and a fractional part neutralised by pure sodium or potassium hydroxide, and distilled in a current of hydrogen. The distillate is neutralised by decinormal acid, each cc. of which is equal to '0014 grm. nitrogen, or '00886 grm. albuminoids. (See also p. 311.) Blank experiments to ascertain the ammonia in the reagents themselves are absolutely necessary. § 147. (7.) Isolation of the Principles Precipitated hy Tannin. — The whey, now free from casein, albumen, peptones, and colouring- matter, but containing mercury nitrate, must be made alkaline, the precipitate filtered off, the liquid saturated with hydrogen sulphide, any precipitate again filtered off", the liquid concentrated to a small bulk and completely precipitated with tannin. The tannin precipitate is decomposed as before described, p. 249. * F. Klug's method of estimating albuminoids by the biuret reaction might proba])ly be applied to milk ; 4 cc. of a solution of albuminoid are mixed with 2 cc. of concentrated soda and 4 drops of 10 ]>er cent, copper solution, and tiltered. The tiltrate is examined spectroscopically (see p. 80, et seq.) ; the absorption factors for X542'5 to X526"7 are as follows: — casein 0"72, albumen 079, peptone 2-65 to 2-635. [Centratbl. /. Phy.uoi., 1893, S. 227.) + ZkUs. Anal, t'hemie, xxii., 366-3S2. §§ 148, 149.] EXAMINATION AND ANALYSIS OF MILK. 289 § 148. (8.) Estimation of Urea. — The estimation of urea is of some importance, since any disorder interfering with the action of the kidneys throws (as has been well ascertained) an excess of urea on all the secretions of the body. A known quantity of milk, which should not be less than half a litre, is evaporated with constant stirring in very large flat dishes to a granular condition. The fat is next extracted by dry ether, and from the fat-free solids the urea is extracted with other substances by boiling alcohol. The alcoholic solution is evapoi^ated to dryness ; and from this dry residue absolute alcohol will extract the urea nearly pure. A litre of milk in its normal state yields about 10 mgrms. Urea must be identified by its properties as follows : — It is crystalline, crystallising in quadratic prisms, and polar- ising with a gentle blue colour under the microscope. The crystals should be heated with a little hydrate of baryta in a closed tube to 200° for some hours, when a very definite reaction ensues, ammonia and carbon dioxide being produced. The liquid may be distilled, and ammonia identified by the Nessler test. Car- bonate of baryta will appear as a precipitate, and may be readily examined, converted into sulphate, and weighed ; 1 part of barium sulphate = -2574 urea. A convenient method of identify- ing urea is also to dissolve the crystals in the smallest possible quantity of water, and then to add a drop of dilute nitric acid ; the nitrate of urea is precipitated, and can be identified by its microscopic characters. Nitrate of urea crystallises on the rhombic system. The most common appearance is that of large plates, many of which lie one upon the other. § 149. (9.) Estii)iation of Alcohol. — A litre of milk, which, if acid, must be neutralised, is placed in a specially constructed, non-tubulated, veiy capacious retort, pi'ovided with a tube a metre in length. This tube is surrounded by a water jacket, through which a continuous stream of cold water runs. The retort tube is pushed through a strong india-rubber stopper, and connected with a small flask holding about 200 cc, and immersed in ice and salt. The india-rubber stopper, by the aid of a second perforation, carries a piece of angle tubing, by which it may be connected with the mercury pump (fig. 4). or, where available, with the ordinary water pump so common in laboratories. The milk is now cautiously raised to the boiling point, and 100 to 150 cc. distilled over. This distillate is redistilled, in the ordinary way, about one-third. All the alcohol is now in a very small compass, and the distillate should be placed in a strong assay flask, and oxi- dised to acetic acid by heating in a water-bath with from 30 to 70 cc. of an oxidising solution of bichromate of potash and 20 290 foods: their composition and analysis. [§§150, 151. sulphuric acid.* The excess of bichromate is reduced by zinc, some phosphoric acid added, and the acetic acid distilled off' by- heating the flask (previously attached to an efficient condenser) over a spermaceti bath. On distilling in this way about three times to dryness, each time adding water, the whole of the acetic acid is obtained in the distillate, and may be determined by titration with a volumetric solution of soda. The amount of alcohol to which the acetic acid found is equal may be calculated by the aid of the following table : — Acetic Acid. Alcohol. 1 -7666 2, 1-53.S2 3, 2-2998 4, 3-0664 5, 3-8330 6, 4-5996 7, 5-3662 8, 6-1328 9, 6-8994 10, 7-6666 The amount of alcohol is of importance in the analysis of decom- posed milks; each part of alcohol may be taken to represent 2-06 parts of sugar, f § 150. (10.) Volatile AciJs. — Volatile acids may be separated in exactly the same way — viz., by careful distillation in a vacuum, first acidifying the milk by tartaric acid. Acetic acid in small quantities is invariably present in fermented milks; but the dis- tillate of normal, quite fresh milk is neutral. In milks already undergoing decomposition, it is best to dilute the milk slightly, and filter through an earthenware cell, and then distil. Under these circumstances, it is not necessary to distil in vacuo. §151. (11.) Estimation of the Total Acidity of Milk — Estima- tion of Lactic Acid. — 10 cc. of d. n. acid are added to 10 cc. of milk diluted to about 100, filtered, the precipitate well washed, and in the clear filtrate the acidity estimated by d. n. soda, using phenyl-phthalein as an indicator; 10 cc. of d. n. soda will have to be subtracted from the total amount of alkali used as representing the acid originally added. Estimation of Lactic Acid. — It is of great importance to make an accurate estimation of lactic acid in milk. A method, accurate enough for technical purposes, consists in thoroughly exhausting * 147 grms. bichromate of potash : 220 grms. of sulphuric acid made up to 1400 cc. with water. t Pasteur, Ann. Chim. Physiqup, Iviii., 323. § 152.] EXAMINATION AND ANALYSIS OF MILK. 291 the milk of fat and lactic acid by etlier, then taking another portion and exhausting it by carbon disulphide, when the difference between the two determinations represents the lactic acid. A more accurate method is as follows : — The milk is dried, exhausted of fat by carbon disulphide, then treated with an alco- holic solution of oxalic acid, filtered, and an excess of hydrated oxide of lead added. Any lactic acid now contained in the fluid will be present in solution as a lactate of lead. The liquid is filtei-ed, saturated with SH^, and again filtered, concentrated by evaporation, and boiled with oxide of zinc; on filtration, evaporation, and standing, crystals of lactate of zinc are pro- duced. There are four isomeric lactic acids ; that which is obtained from milk is fermentation lactic acid, also termed " ethylene lactic acid." The zinc salt has the composition 2(C3Hg03)Zn -f- 3Hr,0. It crystallises in four-sided prisms ; it is soluble in 6 parts of boiling, 58 of cold water. It is nearly insokible in hot or cold alcohol. 100 parts of the salt contain 25'8 of zinc oxide. Lactic jicid itself may be obtained in a very pure state by decomposing the zinc-salt with hydrogen sulphide, when the acid presents itself as a colourless, strongly acid liquid. A drop of this acid, placed in the author's subliming cell (described in the author's work on " Poisons "), and healed very gradually above 200°, gives a white sublimate of lactide, CgH^^Og, a very characteristic reaction. If the heat is not gradual, this sublimate is not obtained, for it then decomposes into carbon dioxide and aldehyde, § 152. (12.) Detection of Metals in ^filk— The detection of the minute qviantities of heavy metals which may occur in milk in cases, where, for the purpose of experiment, salts of the metals have been administered to animals or women, is best conducted by electrolysis, supplemented by the spectroscope. One of the best ways to do this is the method proposed by Dr. Reynolds.* Four to six rather stout platinum wires, half an inch in length, are made into a bundle by binding -svith thin platinum wire, and secured by cotton wool into the throat of the stem of a funnel so tightly, that water placed in the funnel filters through in single drops. The milk, previously acidified with nitric acid, and filtered, is placed in the funnel, and a platinum wix-e, connected with one pole of a battery, inserted in the funnel, so as to be about half an inch distant from the bundle of wire. This bundle is connected with the zinc terminal of the battery- — a single Grove's cell is sufiicient — and the current is allowed to pass until the whole has filtered through. A very decided metallic deposit * Irish Hospital Gazette, 1873. See also " The Spectroscope in Medicine," by C. A. Macmunn, B.A., M.D. London, ISbO. 292 FOODS : their composition and analysis. [§ 153. may be recognised, and examined by ordinary an dysis; but should the deposit be scanty or indistinct, two of the sliort wires are connected with a Ruhmkoff's coil attached to a sufficiently powerful battery. The wires are so adjusted opposite each other as to leave a very short interval between their points, and. the succession of sparks allowed to stream between them is observed by the spectroscope. In this way the spark s})ectra of the metals ax'senic, antimony, mercury, copper, and lead, are easily obtained, and may be identified.* (13.) Detection of Nitrates in Milk. — Pure milk contains no nitrate, hence the detection of nitrates in milk would be fair pre- sumptive evidence of watering. A test recommended by J. Szilasif is as follows: — 1 cc. of a solution of diphenylamine sulphate is ])ut into a small porcelain dish, and a few drops of milk are dropped inj if nitrates ai-e present, a blue colour will gradually appear. Soxhlet tests milk for nitrates as follows : — The milk is coagulated by a solution of calcic chloride free from nitrate, the serum is now treated with a solution of diphenylamine in concentrated sulphuric acid in the same way as with the ferrous sulphate test. The nitrate test will be found of great utility in those cases in which an impure strongly nitrated well-water has been added to milk, but with ordinary waters containing -01 grm. and under of nitric anhydride per litre, the test is not sufficiently delicate to be of much utility. THE MILK SECRETED BY THE UNHEALTHY. § 153. The result of the analyses and cases shortly to be quoted shows, — (1.) That in the case of the cow, in certain diseases only, the milk constantly deviates from the normal standard; (2.) that the most marked changes are found in local diseases of the udder or mammary glands ; (3.) that the animal may be labouring under a most mortal and virulent malady, and yet secrete milk which, although differing from the same milk secreted by the same animal when in health, yet, considered * E,eference may be made to Boisbaudran's "Specti-es Lumineux." Paris, 1874. A simple method of obtaining the wave length value of any spectro- scope scale has been already given in the article on the ' ' Spectroscope " ia the present volume, p. 74, H se^. + Eepert. Anal. Cliemie., 33. § 154.] THE MILK SECRETED BY THE UNHEALTHY. 293 in itself, in no way chemically differs from healthy milk; (4.) that it is only by biological methods of experiment that such diseased milk can be detected. These remarks apply only to the composition of the milk; but if we also regard the quantity seci-eted, then there is in all cases a remarkable difference, for whenever an animal suffers from a sufhcient amount of disease to affect its health materially, the diminution in the total quantity of milk is almost invariable. I. Human Milk. § 154. With regard to the milk secreted by women in various maladies, the same remarks apply only to a certain extent; for the human mammary seci-etion is so dependent on mental influences, that its composition appears readily affected. Vogel gives the following analysis of milk derived fiom a woman suffering fi'om hysteria, the sample being taken directly after the attack: — Per cent. Milk-fat, -514 Casein, 5-000 Sugar, 3-492 Ash, 1-010 Water, 89-984 Specific gravity, 1 -032 Deveux* found the milk of a woman who suffered from nervous attacks, when taken in the seizure, to h& a transparent viscid secretion like albumen. J. F. Simont examined the milk of a woman who was .suffering from the effects of passion. The secretion was apparently the cause of violent con- vulsions and diarrhoea in an infant. The milk was acid, and h id acquired a peculiar odour, and after a little time developed hydrogen sulphide ; or, in other words, the milk had commenced to undergo lactic acid and putrefactive chanoes in the breast itself Local affections of the breast, as might be anticipated, interfere with the healthy action of tlie milk-producing cells. For example, Schlossberger gives the following as the composition of a sample of milk taken from a woman whose breast was considerably enlarged; the fluid was white and thick, and without odour, specific gravity, -98 to -99 atl5°: — Per cent. Fatty matter, 8-54 Lactine and extractives, -75 Casein, 8-74 Ash, -41 The fat fused at :^.3° and solidified at 26°. * Crell's C/ievmche Annalen, vol. 1, p. 369. t J. F. Simon's " Animal Chemistry," Syd. Soc, 11, 53. 294 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 155. II. Cows' Milk. § 155. Aphthous Fever. — One of the few affections in which it is possible for the investigator to discover an abnormal coaclitioa of the milk, and even from the appearance of the fluid to know what particular malady the cow is suffering from, is " foot-and- Tnouth disease" or " aphthous fever J^ This is a febrile, highly infections disease, which has caused of late great ravages among our herds; its most obvious signs are ulcers on the mouth, feet, and teats. Unless the fever is high, the milk is secreted during the whole course of the disease. It presents different (one might almost say, opposite) appearances in different cases; in those ■where there are ulcers on the teat, either externally or just inside, the pus from these ulcers mixes with the milk, and the analyst finds a high fatty residue, from which cholesterin, nuclein, lecithin, and milk-fat may be separated. If, on the contrary, there are no ulcers and no local affection of the vidder, the milk in the more severe cases may be deficient in solids, and especially in milk-fat; nor does it recover its normal composition until about the seventh and eighth days, when the cow begins to improve. A special micro-organism is, according to Klein,* always to be found associated with aphthous fever, it is a micrococcus, which either occurs singly, as dumb bells or as streptococci, the chains of the streptococcus sometimes grow to a considerable length. The individual elements of the diplococci and streptococci are spherical, and have an average measurement of -0006 to -0008 mm. The micro-oi-ganism is readily cultivated in nutrient gelatin, agar-agar or broth; it grows slowly in sterilised milk, the milk remaining fluid. Milk, according to Klein, pi-eserves the micro-organism in a wonderful manner; infected milk kept months in tubes still yielded successful cultures. If the local affection is at all severe, blood-cells, and occasionally a consider- able quantity of blood may be found in such milk. Manimitis. — This disease, as its name imfjlies, signifies an acute inflammation of the mammse. Theoretically, milk secreted by an inflamed organ should be altered much in quality; but, in the case of a heifer suffering from this disease, milk taken the second day after calving did not appear to differ essentially, either in microscopical appearances or in chemical composition, from normal milk. Its specific gravity was 1-0362, and the composition of the solids in 100 parts was as follows: — * Fifteenth Annual Rep. hoc. Gov. Bd., supplement containing report of Med. Officer for 18S5. 155.] THE MILK SECRETED BY THE UNHEALTHY. 295 Milk-fat, 2S00 Casein, ......... 4 "025 Albumen, ......... '560 Milk-sugar, 5 •541 Nitrate of mercury precipitate dried at 100", . . 1'6S* Ash, -920 Chloride of sodium in ash, . . . . . '110 Parturient Apoplexy. — A Coio suffering from Parturient Apo- plexy; Pulse Imperceptible; Temperature %%^-^ F. Third day after calving. Specific gravity, r037. Reaction feebly alkaline. In 100 cc. Milk-fat, 3-750 Casein, 4*025 Albumen, 1-145 (Weight of mercury precipitate, .... 1 -38)* Ash,^ 0-980 NaClinash, 0-102 Urea was absent; there was much lactochrome. No abnormal elements detected by a microscopical examination. The Milk of a Cow suffering from Pneumonia fourteen days after Calving. Pulse 82, Temperature 102°-4 F. Specific gravity, 1 -0297. In 100 cc. Milk-fat, 2-965 Cholesterin, 0-580 Casein, 3-860 Milk-sugar, 3-880 Albumen 0-430 Peptones, 090 Urea, 0-005 Ash, 0-800 NaClinash, 0-488 This is the only milk in which the writer found an estim- able quantity of cholesterin. The microscopical results were negative. * At the time of the analysis the compound nature of the mercury pre- cipitate was not known. 296 FOODS : their composition and analysis. [§ 155. Engorgement of Rumen and Congested Liver. Pulse 68, Temperature 101° F. Specific sjravity, 1'032. ^ * •' In 100 parts by weight. Milk-fat, 6-0-.7 Casein, 4-796 Albumen, 1*067 MUk-sugar, 4-497 Peptones, 0-113 Ash, 0-670 NaCl in ash, 0-092 The milk appears simply concentrated. Phthisis. — A Cow, jive years old, with Extensive Tubercular De- posit in Right lung. The Dam loas also scrofulous. Specific gravity, Milk-fat, Casein, Albumen, Milk-sugar, Galactin, Ash, . NaCl in ash. Dec. 7, 1878. Feb. 1879. 1-0297 1-0340 In 100 cc. In 100 CO. 2-77 3-83 3-650 5-4 0-867 0-365 2 -824 3-34 "> 9 0-866 0-770 096 0-15 A careful microscopical examination could detect no abnormal elements. Phthisis. — A Cow, two years old, in an advanced stage of Phthisis, Jan. 29. Feb. 17. Specific gravity, . 1-0329 10335 In 100 cc. In 100 cc. Milk -fat, . 2-599 3-280 Casein, 3-000 3 980 Peptones, . ? 250 Milk-sugar, . 2-888 4-100 Ash 0-910 0-780 NaCl in ash, 0-10 0-15 The entire amount the cow yielded in January was 1 gallon: the amount sent for analysis was a fractional part of the whole. § 155.] THE MILK SECRETED BY THE UNHEALTHY. 297 A /Sample of Milk drawn from an Udder actually Infiltrated with Tubercular Deposit. Specific gravity, I'OIS. In 100 pnrts by weight. Water, 94-640 Casein, ........ 1210 Albumen, 2-387 Milk-sugar, 0-470 Milk -fat, 0-490 Peptones, absent Urea, ........ 0039 Ash, 0-764 NaClinash, 0-430 Tlie whole quantity of the fluid did not exceed 70 cc. Tt was of a dirty amber colour, with the casein partially separating. A microscopical examination showed very few fat globules, and the following abnormal elements : — 1. Clusters of oval or round granular cells, for the most part '0005 inch in diameter, with a well-marked oval nucleus. 2. Granular masses, irregular in shape, varying in size from about 0-0006 inch to ten or twelve times that size. 3. Granular rounded bodies, stained brilliantly by magenta or carmine. This, then, is phthisical milk in its most intense form, and one never likely to be found in commerce, but admixture of such a. fluid with genuine milk is possible. It is essentially an albuminous serum, containing virea, common salt, and just sufficient casein and milk-sugar to show its origin from a much-diseased milk gland.* Local Affection of the Udder. — MUk from a Heifer two days after calving, suffering from Retention of Fcetal Membrane, a portio7i of the Udder much inflamed. The milk was pink in colour, and contained about a twentieth of its bulk in blood; it was perfectly fresh when examined, but rapidly putrefied. The blood was separated by subsidence as- much as possible. The reaction was feebly acid : — * Possibly the tubercle bacillus will be found in such secretions ; at the time of the analysis of the above, Koch had not published his researches. Good directions for the staining of these bacilli are given in Gibbes' Practical Hi^tolo(jy. London, 1883. 298 FOODS : their composition and analysis. [§ 157. Specific gravity, 1-0313. lulOOcc. Milk-fat, 4-40 Casein and milk-sugar, 9"81 Albumen, 0-62 Peptones, 0-269 Ash, 1-16 § 156. Typhus. — The milk of cows suffering from typhus has been analysed by Husson,* who states that from the commence- ment of the malady, the azotised principles augment, and that there are often found bloody and purulent fluids admixed. The following is an average sample of milk from cows suffering from a not too severe form of typhus : — Fat 1-493 Milk-sugar, 3-140 Albumen, 2-060 Salts, 1-850 § 157. The Propagation of Disease through Milk. — Modern researches on zymotic diseases have for long been converging to the one conclusion, that these diseases are all produced by germs, the life-history of which is analogous to that of bacteria; and that, consequently, such diseases are only special forms of fermentation or putrefaction in living tissues, the disease-zymads growing and multiplying at the expeiise of the tissues. Now, if the composition of milk and of the tissues be com- pared, it will be seen that milk, although physically a fluid, yet resembles in its chemistry a tissue, and contains all that is neces- sary for the nourishment and growth of a zymad. Hence it is, that if a scarlet-fever zymad, or a typhoid zymad, fall into milk, for all practical purposes it is immersed in a tissue ; and cultiva- tion experiments have shown that milk is an excellent medium or soil for the multiplication at suitable temperatures of patho- genic micro-organisms. The Relation of Milk to Scarlatina. — In December, 1885, an outbreak of scarlatina, traced by the author to a particular milk supply, was veiy fully investigated by the Loc. Gov. Board,t and it was proved as clearly as such a matter is capable of proof, that the milk was not infected by any human agency, but that the cows were at the time suffering from a general leverish disorder, the chief external evidence of which was vesicles on the teats and udder rapidly passing into ulcers. The disease is readily propagated by inoculation among calves, and the pathological signs are in many respects strikingly similar to those of human scarlatina. According to Dr. Klein's researches the malady is intimately * Comptes Rendufi, t. 73, 1871, p. 1339. t Fifteentli Annual Report Loc. Gov. Board, containina; Report of Med. Officer for 1885. Ft^: Fv^:Z. *«<«"""" ; / V p * 1 r»^ ' »• »•/ * A / ^ ; ft i! jl 1 k : ^ _-^ Fl^:4. i To Face Page 299, § 157.] THE MILK SECRETED BY THE UNHEALTHY. 299 connected with the multiplication and growth of a micro-organ- ism, which probably gains access to the milk with the discharges from the diseased teats. The micro-organism consists of spherical micrococci, arranged as diplococci, and as shorter and longer, straight, wavy, or curved chains^streptococcus — ^the latter some- times of great length (see Plate, fig. 1). When inoculated into gelatin the growth is slow (see Plate, tig. 2), at the end of a fort- night the streak of inoculation showing up as a white line, made iip of smaller and lai-ger droplets. It does not liquefy the gelatin : it is very similar in appearance and manner of growth to the streptococcus of foot-and-mouth disease, but Dr. Klein thinks the two may be readily distinguished by their action on milk, the foot-and-mouth micro-organism causing no apparent change in sterilised milk; the scarlatinal streptococcus, on the other hand, turning it after two days incubation into a solid mass, A similar organism has been isolated by Dr. Klein from the blood of persons suffering from human scarlatina. Drs. Jamieson and Edington* have also investigated scarlatina by biological methods. Dr. Edington has cultivated the blood drawn from the finger at different dates, and also the epithelial scales thrown off from the skin, taking all possible precautions to avoid aerial contamination. By cultivating a drop of the blood in Koch's jelly, diffusing a minute quantity of this culture through a large body of distilled, sterilised water, and then cultivating drops of this water, the various organisms in the blood were isolated; and from the isolated colonies, pure cultures made; lastly, their effect on animals was ascertained by inoculation. The organism, which was invariably found in the blood of patients during the first three days of the fever, and in the desquamation after the twenty-first day, is described under the name of the Bacillus scarlatince (see Plate, fig. 3). It is in the form of motile rods measuring -4 mm. in thickness, and 1-2 to 14 mm. in length, generally forming very long jointed and curved lepto- thrix filaments; it liquefies the jelly (see Plate, fig. 4). Infected into broth and incubated its growth is very rapid, and a coherent, parchment-like film, which is not easily broken, is formed on the surface ; later the film becomes deeply wrinkled. A calf inoculated with a cu.lture, and also given some in milk to drink, developed fever in six hours, and died within twenty- four. The pathological appearances were those found in the human subject in rapid cases of scarlatina with early death. A second calf was inoculated with a pui-e culture, and developed fever within twelve hours, and there was a vivid rash followed by desquamation ; the throat was severely affected. In six days the temperature was normal. When guinea-pigs and rabbits * Brit. Med. Journal, June 11, 1887. SOO foods: their composition and analysis. [§ loi. ■were submitted to inoculation from cultures, very similar results were obtained. Other organisms isolated from milk, giving either negative or insignificant results, were Sarcina ?w. CO CIC5C533C5 TfipOOlr-aOpP'T-iCI-^ ip ,^, ;-i ^ ^ ,1, CI CI F^ .^ -^ CI CI CI CI CI CI i-0i'*Tf-*r^O— '-HClao p r- --H p CI a> CO -Ti p p ^ioo^o^o-^cib COCOCOTlH-^-^TTtl-^lO OCOdClr-H o-#cicococicicococo >o CO C5 i» CI O I ^ -H C5 00 O O C5 r- OO 00 o o o CO -* lo to cs -H 03 O CO C2 CI CI C< CI CI " ^ -^ ■* . .% g o cS rT ^ bCoT 6 ;^ w ;j ;2,; cT -^ ■^ -^ ^ * +t §1' 333 In the Brit. Med. Journal, July 27, 1895, will be found deter- minations of the milk-fat of seventeen samples of condensed milk ; the following are made from separated milk, and contain from 042 per cent, to 0-96 per cent, of milk-fat : — Marguerite Brand, Tea Brand, Cup Braud, Calf Brand, Gondola Brand, Goat Brand, Swiss Dairy, Wheat Sheaf Brand, Clipper Brand, Daily Brand, Cross Brand, Shamrock Brand; the following from 1 per cent, to \-b per cent.: — Home Brand, Handy Brand. The Nutrient Brand has 2-36 per cent.. As you like it Brand, 4-23 per cent. In the Analyst, July, 1893, will be found a useful paper by Messrs. Droop Bichmond and L. K. Boseley on points on the analysis of condensed milk. KOUMISS. § 177. Koumiss is an alcoholic drink made by the fermenta- tion of milk ; it is prepared by the nomad population of Asia (especially by the Tartars) from the milk of the mare and that of the camel : it is also manufactured from cows' milk. The preparation of koumiss by the Tartars is very simple : ten parts of fresh wai-m milk, with a little sugar, are added to one part of milk which is already sour — that is, which contains lactic ferment — and the whole is allowed to I'est for two or three hours with repeated stirring. The chemical changes taking place seem to be a partial decomposition of the sugar into lactic acid, the development of carbon dioxide and alcohol, and possibly certain changes in the albuminoids, changing them partly into peptones. The composition of koumiss, since it may be derived from such different sources, is variable. A few analyses are as follows : — Mean of Koumiss from Komniss from Koumiss ten analyses. mares' milk. cows' milk. 48 hours old. Koaig. W.Fleischmann W.Fleischmann J. A. Wanklyn. Water, 87-88 91-53 88-93 87^32 Milk-sugar, 3-76 1-25 3-11 ) 6 •GO Lactic acid, 1-06 1-01 •79 Casein, 2-83 1-91 2-03 2^84 Milk-fat, . •94 1-27 •85 •68 Alcohol, 1-59 1-85 2-65 1-00 Carbonic acid, •88 •88 1-03 -90 Ash, . . . 1-07 •29 •44 •66 In the koumiss from cows' milk, Fleischmann separated '166 per cent, of glycei-in. 334 FOODS : their composition and analysis. [§ 178. LEGAL CASES RELATIVE TO MILK. § 178. The following convictions are of interest : — Appeal Case, in which the defence was, that the milk had been deprived of cream from being unintentionally skimmed by serving the customers. This case occurred at the Liverpool Sessions, and is fully reported in the Analyst, 1877, p. 214. Dr. Campbell Brown proved that the milk was deprived of its cream, the appellant affirmiug that the milk had been put into an eighteen-galloa can, without the slightest sophistication, and that the cream had been abstracted by serving the customers. The Recorder said, Nobody would convince him that a milk-dealer could not, if he liked, take care that each of his customers should get a fair pro- portion of cream. ... He was perfectly certain that the milk had not been skimmed, but that it had been weakened by the process of selling to the earlier customers. He was certain that when the ajtpellant sold the milk to the earlier customers, he knew he was abstracting the cream from it— not skimming the milk, but abstracting the cream, although with no fraudulent intention. He was equally certain, too, that the ajipellant sold the residuum of the milk, knowing that it had been reduced to the condition in which it was when he sold it. He was quite satisfied, therefore, that an offence had been committed against the Act of Parliament. . . . The conviction was confirmed. The officials at Somerset House have, in an appeal case, also declared their belief in the fact that milk on being served from a can, in the usual way, may have the top layer of cream entirely abstracted. The writer has always had doubts about this, for the difference of specific gravity between the cream and the solution of the other milk solids is not great, and the mere use of a dipper in serving the milk stirs it up sufficiently to render this removal improbable. We have, however, at least, one definite experi- ment on this subject. Mr. Carter Bell states,* "One day in July I bought two gallons of milk, and analysed it, and found 100 cc. to have the composition of — Total solids, 12-30 Fat, 2-70 "Solids not fat," 9-60 The milk was put in the cellar, and at every hour from nine in the morning till twelve o'clock at night, one pint of milk was taken out at the commence- ment of each hour, and a portion of each pint was analysed. In taking out the pint, great care was taken not to stir the milk, the measure was simply dipped into the milk and taken out. The whole experiment was conducted throughout in favour of the milkman, and according to these experiments it is more advantageous for customers to be late than early." 8 o'clock in the morning, 9 10 11 12 • Analyst, No. 21, December, 1877, p. 162. Total solids. Fat. 12-30 2-70 12-68 3-08 12-68 3 08 12-70 3-10 12-70 3 10 178.] LEGAL CASES RELATIVE TO MILK. 335 Total solids. Fat. 1 p.m., 12-24 2-64 2 „ 12-30 2-70 3 , 12-28 2-68 4 „ 12-88 3-28 5 12-80 3-28 6 „ 12-40 2-80 7 12-54 2-94 8 „ 12-30 2-70 9 , 12-48 2-88 10 „ 12-88 3-28 11 „ 12-60 3-00 12 , 12-90 3-30 Manufacture of New Milk from Condensed. The only interest of the following case lies in the revelation it aiFords of the tricks of the trade. A man was summoned by his employer, a dairyman, for adding dirty water to milk. The prisoner did not deny the accusation, but cross-examined the prosecutor, to show that the latter had been in the habit of making his men, when the milk ran short, mix with water a quantity of white stuif that was kept in the cellar, and take out to the customers to make good the deficiency in the supply of good milk. The process he called "the fake" of the trade. The prosecutor admitted that he kept condensed milk to make up the supply when the demand was too great, the "white stuff" referred to by the prisoner.* Novel Defence. A defence was set up, in a Swansea case, that the poverty of the milk was owing to its having been taken from a cow a few hours only after she had been milked dry. Mr. Morgan instituted experiments on this point. In eighteen experiments on the same cow the following results were noted : — t Total solids — Highest, 17-60 Lowest, . , . . . . 12-59 Average, 13 93 Fat- Highest, 8-60 Lowest, 2-96 Average, 4-41 Solids not fat — Highest 9-95 Lowest, 9-00 Average, 9-52 * Analyst, iL, 1878, 184. t Proceedings of Society of Public Analysts, i., 1876, p. 191. 336 foods: their composition and analysis. [§ l"'^- Adulteration of Milk with Cane Sugar and Water. At the Colomb Petty Sessions, in Jan. 7, 1879, a niilk-dealer was sum- moned for the above offence. The composition of the milk was : — Water, 87-63 Milkfat, 3-00 Casein, 2-90 Cane sugar, 2-80 Milk-sugar, 3-10 Ash, -37 The summons was dismissed on the ground (which has since become unten- able) that the purchase being admittedly for analysis and not for consump- tion, the purchaser was "not prejudiced." — Analyst, 1879. Defence that Bain had increased the quantity of Water. In December, 1880, a cowkeeper of Hull was summoned for selling milk adulterated with 10 per cent, of water. The defendant affirmed that whilst milking the cows in the field on the morning in question, it rained very heavily, and he thought that about a pint of water fell into each of the milking buckets. The court did not consider the defence valid, and con- victed and fined the defendant. — Analyst, 1880. Conviction for selling " Fo7-e^' Milk. In August, 1S77, a dairy proprietor of Dublin was prosecuted for selling milk deprived of its cream. The defendant stated that it was " fore" milk, and that he had sold the " strippings " as cream. The magistrate expressed his opinion that the milk should be sold whole, — i.e., with both "fore" milk and " strippings," and fined the defendant £10.* Diseased Milk. At the ^Yoolwich Police Court, in December, 1875, a daii-yman was con- victed and fined £20 for selling diseased milk. Mr. Wigner, the analyst, proved that the sample had a peculiar colour, and that it contained no less than 13 per cent, of fat, 8"2 "solids notfat," and 20 per cent, of blood. Other witnesses proved that the defendant had a number of cows, and at least one of them was suff'ering from foot-and-mouth disease. There was practically no defence, t Appeal Case.— Conviction that Milk containing 2'69 jnr cent, fat is not of the Substance, Nature, and Quality demanded by the Purchaser. — Times, Nov. 7, 1893. This important case was an appeal heard before Queen's Bench (Mr. Justice Charles and Mr. Justice Wright). An Inspector procured a sample of milk from a churn of milk in course of delivery at a railway station. The churn was labelled "the contents of this churn are warranted to be new and pure milk." On analysis, the total solids were found to be 11 per cent., the fat 2-69 percent. The magistrates convicted. The defendant appealed. It was stated that the milk was from a dairy of twenty-six cows, that the appellant had not removed cream from the milk, and the * Analyst, i., 1877, p. 82. t Conviction for Selling Milk yielded by a Cow suffering from Disease. London, 1876. § 178.] BIBLIOGRAPHY. 337 appellant ascribed the poorness of the milk to the dryness of the season or to the quality of the food, and it was contended that there was no evidence of actual adulteration, and that the poorness of the milk might be from natural causes. Mr. Poland, Q.C., urged that the magistrates had not found as a fact that any cream or fat had been abstracted from the milk, and that selling poor milk was no offence, unless it was so by adulteration or withdrawal. Mr. Justice Charles in giving judgment stated that the result of the analyst's certificate was that the proportion of cream or fat in the milk was less than natural or usual, and that it showed an offence against the statute. Mr. Justice Wright concurred, and the appeal was accordingly dismissed and the conviction upheld. BIBLIOGRAPHY. The general treatises enumerated at p. 43, all contain articles on milk, and may be consulted. The limits of this work pre- clude an exhaustive bibliogra])hy on milk — a list of the more important papers which have been consulted in the writing of tlie sections on milk would alone occupy more than a dozen pages. Hence it will be necessary to limit the list to the papers wltich have ajipeared since 1870. Abstract of Work^done by the Milk Committee. Analyst, 1886, 2, 62. Ad.am, a. F. — "Etude sur les principales methodes d'essai et d'analyse du lait," &c. Paris, 1879. Adams, M. A. — On a new Method for the Analysis of Milk. Analyst, 1S85, 46. The Treatment of Sour Milk intended for Analysis. Analyst, 1885, 99. Allen, A. H. — Note on the Optical Estimation of Milk-sugar. Analyst, 1885, 72. The Preservation of Milk Samples for Reference. Analyst, 1886, 203. AxMAXN. — Ueber Milch-Verfnlschung und Milch-ControUe. Zurg. Cor- respond. BL, 1873. Ballard. — 3Ied. Times and Gazette, 1876. Bell, J. C. — Milk Analysis. Analyst, December, 1877. Besana, C. — Sheep's Milk, and the Manufacture of Cheese therefrom. Staz. Sper. Ag. Ital., xxiii., 572. Blyth, a. Wyntek. — Milk of the Cow in Disease, Chem. News, vol. xxxii., p. 245 ; Analyst, i., 1878 ; ib., 1879, 231. Milk of the Cow in Health and Disease, Journ. Chem. Soc. , 1 879. Exj)eriments with the Lacto- crite. Analyst, 1887, 34. Identification and Estimation of Carbo- hydrates in Milk. Analyst, March, 1895. Bollinger, O. — Ueber Impfs-u. Futterungs-tuberkidose, Archiv fur Expert- mentelle Pathologie u. Pharmakologie. Leipzig, 1S73, 356-375. Bollinger, 0., u. Franck, E. — Gegen die grosse (iefahrlichkeit des Genusses vom Fleisch u. Milch tuberkuloser Thiere. Deutsche Zeitschrift fur Thiermedizin u. vergleichende Pathologie, 1875, t. 1. BoucHARDAT, A. , et T. a. Quevenne. — "Instruction sur I'essai et I'analyae du lait," 3. ed. Paris, 1889. 23 338 FOODS : their composition and analysis. [§ 178. Cameron, A.— Analysis of the Milk of Forty-two Cows, Analyst, 1881. Embrey, G.— The Lister-Babcock Milk Tester, Analyst, May, 1893, EsTCOUKT, C— On the Valuation of Milk Solids, Analyst, 1883, 245. Faber, H.— a new Method of Determining Fat in Milk, Analyst, 1887, 6. Condensed Milk, Analyst, 1890. Fleischer, M. — Journal f. Landiv., 1871. Hehner, 0.— Condensed Milk, Analyst, vol. iv., 1879. Heisch, C — On Diseased Milk, Analyst, 1878, p. 249. Hill, A. — Milk Standards, Analyst, May, 1876. Hill, Eustace. — The Werner-Schmidt Process of Fat Extraction, Analyst, April, 1894. HoPPE-SEYLEK.~-"Physiologische Chemie," Berlin, 1877-80. "Handbuch der Physiologisch. und Pathologisch-chemischen Analyse," Berlin, 1875. Jamieson, W, Allan, M.D., and Edington, Al. — Observations on a Method of Prophylaxis and an Investigation into the Nature of the Contagium of Scarlet Fever. Brit. Med. Journal, June 11, 1887. Johnstone, VV. — The use of Acetic Acid in Milk Analysis. Analyst, 1886, 22. Kehrer. — Centralb. /. die Med. Wiss., 1870, No. 33; Arch./. Gynakologie, ij.. Heft., i. and iii. Klebs, E. — Die kiinstliche Erzeugung der Tuberculosa, Archiv fiir experi- mentelle Fathologie u. Pharmakologie, 1 Band. 2 Heft, Leipzig, 1873, s. 177. Klingler. — Dingr/er's polytechnisch. Journal, 217, Augsburg, 1875, s. 343. Kuhn, G. — Sdchsische Landui. Zeitung, 1875. KuLS, J. — PJliiger's Archiv., xiii., 176-196. Leffmann, H. and Beam, W.— Analysis of Milk and Milk Products. Philadelphia, 1893. Liebermann, Leo.- — Ueber in der Milch enthaltenen Eiweiss-Korper, Zeits- chrift f. anal, Chemie, xxii., 232. Volumetrische Methode zur Bestimmung des Fettgehaltes der Milch, Zeitschrift f. anal. Chemie, xxii., 383. Macadam, Dr. Stevenson. — The Quality of Milk supplied to Towns, Pharm. Journal. Macnamara, F. N.—Chein. News, 1877. Milch-Zeilung—Y&r'was Papers on the Composition of Milk, &c., &c., 1870-80. MtJLLER, A..— Zeitschrift f. anal.Chem., 22, 129. Muter, J. — An Ingenious Adulteration of Milk, Analyst, 1, 1878. Paris, C. van. — Uebertragung der Stomatitis Aphthose der Kinder auf den Menschen durch Milchgenuss, J. Sch. Jahrbuch, 1873. Pattinson. — An Abnormal Sample of New Milk, Analyst, 1876. Pavesi, a., ed E. Rotondi. — Gazzetta Chimica Italiana, Iv., 1870, 194. Pfeiffer, Emil. — Zur Quantitativen Analyse der Muttermilch nebst einera Auhange Uber Kuhmilch. Zeits./iir anal. Ckern., xxii., 14. Redwood, Dr. — " Notes on the Analysis of Milk," 1, 1876. Richmond, H. Droop. — Preservation of Samples of Milk, Analyst, 1890. Fat Extraction from Milk Solids, Ibid. Composition of Milk and Milk Products, Ibid. ; also Analyst, March, 1895. Richmond, H. Droop, and Boseley, L. K. — Points in the Analysis of Con- densed Milk, Analyst, July, 1893. Note on the Detection of the Adulteration of Fresh Milk by diluted Cow's Milk, Analyst, July, 1893. Richmond, H. Droop, and Pappel, A. — The Milk of the Gamoose, J. Chem. Hoc, Trans., 1890, Ivii. , 754. § 178.] BIBLIOGRAPHY. 339 RiDEAL, S. — On Formalin as a Milk Preservative, Analyst, Jul.y, 1895. RiTTHACsEX, H. — Neue Methode zur Analyse der Milch und iiber ein von ]Milchzucker verschiedenen Kohlenhydrate der Kuhmilch, Journal/. prakt. Chemk, 1S77. Selmi, F. — GazzeUa Chimica Italiana, iv., 482-484; Journal Chem. Soc, Lond., 1875. Smee, J. H. — " Milk in Health and Disease." Lond., 1875. SoxHLET, F. — Beitrage zur physiologisch. Chemie der Milch, Journ. f. prak. Chem. 1872. Die chemischen Unterscheide zwischen Kuh u. Frauenmilch u. die Mittel zu ihrer Ausgleichung. Munich, 1893. Stevenson, J. — The Decomposition of Milk, Analyst, 1, 1876. Stokes, A. W. — On the Determination of Mixtures of Milk-sugar and ■ Cane-sugar, Analyst, 1885, 62. The Werner-Schmidt method of Fat Extraction, Analyst, April, 1894. Struve, H. — Studien iiber Milch. Journal f. praktische Chemie, xxvii., 249; xxviii., 70, 110. Thomson, W.— The Incongruity of the Mode of stating Milk- Analyses, Analyst, Sept., 1877. Vallin, E.— Le Lait des Vaches Phthisiques, Ann. d' Hygiene Publique, 1878. ViETH. P.— Milk and Milk Products, Analyst, 1885, 67 ; 1886, 66; 1887, 39. The Polarisation of Milk-sugar, Analyst, 1886, 141. Easy Methods for the Examination of Milk. London, 1887. Fat Extraction and Estimation in Milk Analysis, Analyst, Nov., 1891. Composition of Milk and Milk Products, Analyst, April, 1891, and April, 1892. Werner. — Method of Fat Extraction. Zeit. f. anal. Chem., xxvii. 340 foods: their composition and analysis. [§ 179. BUTTER. § 179. Constituents of Butter. — In the manufacture of butter the cream is violently agitated in a churn or other suitable appa- ratus, and in this manner the thin membrane* enclosing the fat globules is supposed to be ruptured. The free fat then coalesces, entangling with it some casein and serum ; the butter is well pressed together to free it as much as possible from moisture, and salt added to assist its preservation. Butter, therefore, is com- posed principally of milk-fat, with a small and variable quantity of water, casein, and ash, the latter consisting chiefly, but not entirely, of the salt added. The "/«'" of butter may be shown, by careful cooling, to consist of about 45 '5 per cent, of butter oil and 54 "5 per cent, of solid fat ;+ it is usually stated to consist of a mixture of the glycerides of the fatty acids — palmitic, stearic, and oleic — not soluble in water ; and also of the glycerides of certain soluble and volatile fatty acids, principally butyric, with small quantities of caproic, caprylic, and capric acids. It is the association of about 7 "8 per cent, of the triglycerides of these volatile acids with the glycerides of the insoluble acids, which gives to butter-fat its peculiar and distinctive characters ; but it is probable that stearin, palmitin, butyrin, and caproin do not exist in butter, their place being taken by more com- plicated glycerides, the glycerin being combined with two or three different /O.C4H7O acids. A crystalline glyceride,^ CsH^^— . C1CH01O2, has, indeed, been isolated from butter. \0 . C18H35O2 The dift'f^rent constituents, as well as the physical character- istics, of butter- or milk-fat have been already described at p. 239 et seq. The general composition of butter-fat, as usually stated, is as follows : — Glycerides Equal to Fatty Acids. Olein, . .42 21 Stearin and Palmitin, . 50 00 = Oleic Acid, r Stearic and Pal- 40-40 ~ ' \ mitic Acids, . 47-50 87 -90 Total insoluble solids. Butyrin, . 4-67 = Butyric Acid, . .S-49 Caproin, . 3-02 = Caproic, . .2-40 Caprylin and 1 .,^ _ f Caprylic and Pvutic Paitin, . j ^" - 1^ Acids, . . -80 100 'OOg = 04-59 Total acids. !l Pure, dry butter-fat, melted at a heat not exceeding 100° Fahr., * Reasons for doubts as to the existence of this membrane are given at page 237. t A. Wynter Blyth and Pobertsoft, Joiirn. Chem. Sor., 1889 (Proceed- ings), 5. X A. Wyter Blyth and Piobertson, Op. cit. § The theoretical percentage of C, H, and 0, corresponding to these glycerides, is as follows :— C 728, H 13-3, 13-9. II E. Duclaux (Compt.'Rvnd., cii., l,0-2'2, 1,077) by a process of fractional distillation has examined a series of prize Normandy butters, in special §180. OLEO-MARGARINE — EUTTERINE. 341 has at that temperature a specific gravity ranging from -91079 to -91400 ; its fusing point, taken in the manner to be described, ranges from 30°-5 to 36°-5 ; average specific gravity at 15° is •93072. The relative proportions of fat, casein, and salt, in genuine butters, may be gathered from the following table, in which it is seen that the butter-fat ranges from about 82 to nearly 87-5, the average given by Angell and Hehner being 85-45 per cent. : — Normandy Butter. Angell and Hehuer. A Sample of Fresh Butter. Angell and Hehner. Butter from Isle of Wight. Angell and Hehuer. Butter Guildford. Angell and Hehner. Butter from Win- chester. Angell and Hehner. Mean of Eighty. nine Analyses. Konig. Fat, .... Curd, .... Salt (Ash), . . Water, . . . Milk-sugar, . . 82-643 5137 2-915 9 305 83-871 i 84-740 2-721 3-462 0-424 2 089 12-984 9-709 85-480 2-789 3-151 8-580 87-223 2 054 2-108 8615 83-11 •86 1-19 14-14 •70 OLEO-MAEGARINE-BUTTERINE. § 180. Oleo-margariue, Dutch butter, butterine, edible fat, and similar appellations, all denote difl:erent varieties of a manu- factured article made expressly to imitate butter.* It is an im- portant industry in the States, and is also largely pi-oduced in Holland (hence the name of Dutch butter) and in Belgium. As manufactured in Chicago, it would appear to be, at all events, a relation to the proportions of Caproic and Butyric Acids; his results are contained in the following table ; — Water, . Fat, . Milk-sugar, Casein and f Caproic Acid (per cent.), Butyric Acid (per cent.), Sum of the acids, . Eatio, . 10000 2 10 3-55 5-65 21 10000 2-18 .3-52 5-70 20 10000 217 3-50 5 70 2-0 10000 2-23 3-60 13-34 SG-01 0--20 1162 86 -02 o-?o 1-56 14 00 8.5-31 0-20 0-49 100 00 2 08 3.52 100-00 2-19 3 46 The mean of the eight determinations gives Caproic 2^15 and Butyric Acid 3-52 per cent. ■* Mc'ge Mouries, a chemist, appears to have been the first who proposed the manufacture of artificial butter. It came first into commerce about the year 1872. 342 FOODS : their composition and analysis. [§ 180. cleanly article. Animal life is plentiful in the States; American technical operations are on a gigantic scale, and too much capital is involved in the matter to allow of the use of any process likely to disgust the consumers. The chief constituent used is beef-fat, which consists for the most part of stearin, margarine (so-called), and olein. The olein and margarine melt at a much lower temperature than the stearin. Mutton-fat contains more stearin than beef-fat ; hence, in summer, the softness of beef- dripping as compared with the solidity of mutton-fat. This is obviously the reason why the manufactui'er prefers beef-fat. The l^rocess of manufacture is briefly as follows : — The beef-fat, freed first as much as possible from fibre, passes in a very finely- divided state from a sort of mincing-machine, technically called a "hasher," to large tanks, where it is melted by means of water- jackets ajjplied to the tanks, and heated to a temperature never allowed to exceed 39°. The result of this process is, that the fat melts to a clear yellow oil, the water and debris sinking to the bottom, and a thin scvim of impurities rising to the surface. The latter is skimmed ofi', and the yellow oil run into wooden cars, in which the stearin, after a little time, begins to deposit in a more or less crystalline or gi-anular condition ; the refined fat is then put in a press-room, and kept at a temperature of from 2G°'6 to 32°'2. The oleo-margarine is filtered through cotton cloths, and ultimately pressed ; the result of which is, that the stearin is left behind as a white cake, and is ultimately disposed of to the candle-maker. The oleo-margarine, at this stage, is quite tasteless, and has no fiavour of butter. This fiavour is given by churning it with milk ; lastly, the product is coloured with annatto, and rolled with ice, after which it is either made up into pounds, or packed into kegs for export. Arrived in this country, it is either sold honestly as " margarine," at the price of about a shilling per pound ; fraudulently at a higher price, as butter ; or it is used as an adulterant of butter. The chemical proportions of the artificial butter vary according to the fats used in the manufacture and the details of the process employed ; but they all agree in this, that when the butter is saponified and the acids set free, there is a great deficiency of soluble fatty acids, as compared with those of true butter-fat. The average percentage composition of dry margarine-fat is as follows : — Palmitin, 22-3* Stearin, 46-9 Olein, 30-4 Butyrin, Caproin, and Caprylin, . . '4 * Hence a combustion of margarine-fat gives higher numbers for carbon than pure butter fat, the theoretical amounts for the above being — C 74 "6, H 13-2, 12-2 per cent. § 181.] ANALYSIS AND ADULTERATION OF BUTTER. 343 While the proximate analysis of commercial margarine is as follows : — Water, 12 01 Palmitin, 18-31 Stearin, 38-50 Olein, 24-95 Butyrin, Caproin, and Caprylin, . . '26 Casein '74 Salts, 5-23 • ANALYSIS AND ADULTERATION OF BUTTER. § 181. The only common adulterations of butter are the sub- stitution or admixture of fats other than butter, of water,! the latter being either left in the butter in undue proportion through faulty manufacture, or fraudulently added, of colouring-matters, and of boracic acid or borax. The addition of mineral sub- stances,! flour, and other articles enumerated by diiferent writers, is at the present day rare. The analysis of butter naturally divides itself into — (1.) The general examination and analysis; and, (2.) The investigation of the fat. 1. The General Exaviination and Analysis of Butter.— The colour, taste, and odour of the sample should, of course, be noted. It will also be found useful to examine it in thin layers microscopically. If it has been mixed by fusion with any fat, and cooled slowly, crystals may be discovered. The best way to seek these crystals is to place a minute portion of the fat on a slide, add a drop of castor or olive oil, press the thin disc of covering glass so as to get a very thin layer, and examine by polarised light. Under such circumstances, if crystals should be present, there will be seen dark crosses similar to those in potato starch. Such crystals are suspicious, because they show that the butter has been melted ; and it certainly must be a most unusual process to melt butter save for the purpose of mixing with other fats. The rare adulteration of any other substance, * It has been asserted that artificial tributyrin has been added to oleo- margarine. Such an admixture may be readily dissolved out of the butter- fat by strong alcohol, while the natural tributyrin is not so easily separable from butter-fat by any solvent. t A. Mayer [Lmidw. Versuchs.-Stat., 29, 215-232) has made experiments on a man and a boy as to the relative nutrient power of natural and artificial butter. He found on the average that TG per cent, less of the artificial butter was absorbed than of the natural. X Silicate of sodium has been found in modern butters, and must be looked for, see p. 345. 344 FOODS : their composition and analysis. [§ 181. such as starches, &c., by mechanical admixture, cannot fail to be detected by the microscope. The Proximate Analysis of Butter — that is, the separation of butter into mineral matters, casein, butter-fat, and water, is very readily performed. 10 to 20 grms. of the butter are weighed into a counterpoised porcelain dish, and melted over a low gas flame, keeping the butter at any temperature between 105° and 110°, with constant stir- ring, until all effervescence has ceased. (By using a thermometer as a stirrer this is easily effected.) The weighing of the dish and its contents when cool, subtracted from the first weight, gives the loss ecpialling the water, and it may be worked into per- centage. Keeping the butter for a considerable time at 100°, or above, must be avoided, for it increases the weight. The above method has the merit of expedition, and it is tolerably accurate ; but the water may also be estimated by placing about 1 grm. of the butter in a large platinum dish, so that the fat forms a thin layer, and then exposing it to tlie heat of the water-bath until it ceases to lose weight. The salt is best determined in a separate sample ; from 5 to 10 crms. of the butter, made up from different portions of the sample, are shaken up with hot water in a separating funnel, the water poured away from the fat, and the chlorine estimated by silver nitrate and potassic chromate, as described in the article on "Water Analysis. The fat is estimated from the dried butter, and the melted fat is poured off, as far as possible, clear from the curd and salt, the residue being thoroughly exhausted by boiling benzine, ether, or petroleum, which can be effected, if care is taken, in the same dish without transference to a filter. On now weighing, the loss equals the fat. Lastly, the curd is burnt away at a low red heat, and the ash weighed. The general analysis finished, it remains to consider the results : — (1.) Fat. — The fat should not be below 80 per cent. ; any figure under this should justly be considered evidence of adul- teration. (2.) Water. — There is no standard followed or fixed with regard to the percentage of water. In those cases in which the fat is below 80 per cent., the deficiency of fat is usually from excess of water ; and seeing the variable quantity of water found in butter, it is wisest not to certify on the ground of water alone, unless there is sufhcient to lower the percentage of fat below 80. (3.) Caseifi. — The average quantity of casein is 2*5 per cent., § 181.] ANALYSIS AND ADULTERATION OF BUTTER. 345 but it may reach 6 to 7 per cent., and the higher the percentage of casein the less likely is a butter to keep, although this usually is evidence of error in the manufacture rather than of adulteration. (4.) The Ash. — This should consist of common salt and phos- phate of lime. Butter is said to be adulterated occasionally with sodium silicate, and therefore the ash should be fused with sodic carbonate, dissolved in hydrochloric acid, evaporated to dryness, and dissolved in water. Any residue will consist of silica. If other mineral adulteration is suspected, a complete analysis of the ash may be necessary (see p. 119). There is no definite standard fixed with regard to the weight of the asli, but most chemists agree that it should not exceed 8 per cent. For the detection of borax in the ash, see the article on ]\Iilk (p. 311). The following are examples of adulterated butters, the adulter- ation being detected simply from the proximate analysis : — Devon Butter. Devon Butter. A Sample of Butter. (Angell and Hehner.) A Sample of Butter. (Angell and Hehner.) Fat, Casein, . Water, . . Salt, 78-50 1-72 1710 2-68 76-34 6-60 13-36 3-70 67-580 6 -880 23-981 1-559 47-019 7-854 42-358 2-689 Colouring-Matter of Butter. — The colouring-matter of natural butter is the " lactochrome " already described (p. 248), but a great many, otherwise genuine, butters are coloured with the harmless colouring material " carotin." A method of detecting carotin has been described by Mr. W. Moore. "^ 10 grms. of the butter-fat are dissolved in just sufficient CSg to which 20 cc. of absolute alcohol are added, and the mixture shaken vigorously. If carotin is present, the CS., is coloured, but the alcohol is colourless ; on the addition of a drop of ferric chloride, this condition is reversed, the CS., becoming colourless, and the alcohol layer coloured.! Most of the artificial colouring-matters can be washed out by alkaline water from an ethereal solution of the butter-fat. •* Analyst, Sept., 1886. t Butters are also coloured with annatto (see p. 100), turmeric, saffron, marigold leaves, and yellow wood. 346 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 182. 2. Exainination and Analysis of the Fat. — By far the most important process in butter analysis is the examination of the fat. The data by which the analyst judges whether a butter consists of foreign fats, entirely or partly, are derived from — (a.) Certain simple tests ; {b.) the melting-point; (c.) the angle of refraction ; {d.) the specific gravity ; (e.) viscometry; (/) acid value; {g.) Koettstorfer's test; (Ji.) the relative proportion of soluble and insoluble fatty acids ; (i.) Reichert-Meissl's distilla- tion process. For many of these tests the first requisite is a pure dry fat. This is easily accomplished by melting a sufiicient quantity of the butter over the water-bath. In a short time the water, curd, and salt sink to the Ijottom, and the nearly pure fat can be poured off. Should it not be clear, it must be filtered through filtering paper or glass-wool. This operation will necessitate the filter being kept warm in a suitable steam-jacket. § 182. (a.) Certain /Simple Tests. — In the winter of 1880 the author made various experiments on butter-fat, and dis- covered the curious fact that all solid fats, when melted and dropped on to water, the temperature of which is low enough to ensure their solidification, set in a definite form or pattern. To attain success, it is necessary that the fat as well as the water be of a certain temperature ; but with many of the glycerides and mixtures of glycerides, such, for example, as butter, butter- ine, dripping, &c., the range is very wide; so that if the fat is perfectly fluid, and within one or two degrees of 100°, the water ranging from 0° to 15°, a pattern more or less perfect is obtainable. Each fat appears to have its own distinctive pattern, and can be identified by its pattern alone. On the other hand, each fat has a variety of patterns, for every alteration of the experi- mental conditions modifies more or less the form of congealed drops. If, however, the conditions under which each experiment is performed are precisely similar, there is no difliculty in obtain- ing the same form, or very similar forms, any number of times. The chief modifying conditions are the difference of tempera- ture between the fluid fat and the water, and the height from which the fat falls. I have found that from 3 to 4 inches is the best height, and that a greater fall than this tends to spread the films out, and renders all patterns more or less similar. Referring to individual forms — Butter. — The experiments were made on several samples of butter, whose genuineness had been proved by analysis. The fat was melted and filtered, and kept in an air-bath at tempei-a- tures of from 40° to 80°, and then dropped from a clean warm glass rod on to water of from 10° to 15°. The most common and distinctive form attained in this way was that of a beautiful No. 1 . TALLOW. No. 2. SPERMACETI DROPS OF TALLOW AND SPERMACETL § 182.] ANALYSIS AND ADULTERATION OF BUTTER. 347 foliated film, not unlike the leaf of a pelai-gonium. The film may be transferred direct to the lithographic stone, and one may thus have a direct impression and a permanent recoi'd. The details of delicate veining are, as might be expected, lost. The best pattern temperature for butter is 55°, the water being at 10° ; but regular forms may be obtained up to 100°. At higher temperatures success is rare. I found that although butter of 40° to 50° when dropped on to water of 10° sets in a radiated star form, yet when dropped on to water of 8°, although momentarily there was a beautiful complicated foliation with many radiating wings, these wings suddenly mutually repelled each other, and the pattern fell, or rather flew to pieces. Glass plates were prepai'ed chemically clean by first treating with alcoholic soda and then washing them with ether ; the plate was next dipped into water, and thus a thin water film obtained. On this perfectly smootli wet surface, butter and other fats were dropped. In the case of butter, the pattern lost much of its beauty, but was always very regular in outline. Butter films ai'e of extreme tenuity, and although several attempts to photograph them were made, the light passed through almost as perfectly as through glass, there- fore the photographic shadows were too indistinct to make any use of. Margarine. — The various mixtures of animal fats in the market known as margarine, or artificial butter, give by no means iden- tical patterns, for they vary much in composition ; but in each case the foi'm can be distinguished from the butter films, and from the pattern alone it is always possible, and often very easy, to say whether a given film is butter or not. The best method to distinguish the artificial from the genuine product, is to take pure butter-fat and the suspected sample, and after melting them each at the same temperature, to drop them on to the same glass plate side by side : the margarine pattern is full of minute crystals ; the butter i)attern has no crystals. The pattern of one sample of margarine was found to be identical with that of tallow — little white dots containing bunches of crystals (see the photograph, No. 1, Tallow). Tallovj. — One form of tallow pattern has just been alluded to. It is the most common form when melted tallow falls on water. This fat of high melting point is a good illustration of the many forms which may be produced at diff'erent temperatures. Thus, at 0° the fat sets on water in circular indistinct drops, but when more fluid, and dropped on glass, its pattern is distinct and crystalline. Paraffin, giving no pattern by itself, when mixed with other animal fats, as may be expected, profoundly modifies the film. 348 FOODS : their composition and analysis. [§ 183. Eqiial parts of paraffin and stearic acid, for example, give by suitable treatment a pattern in the shape of a broad cross. With reference to the fat patterns of spermaceti, stearic acid, and generally fats of high melting points, it need scarcely be said that it is impossible to obtain them by dropping the fat on to cold water. Such a proceeding only gives a shapeless mass. To be successful it is absolutely essential that the water of the glass plate should be warm. For example, spermaceti gives no definite form when melted at 100°, and dropped on to cold water, or even water of 50° ; the water must be heated up to 80° or 90° for a good result to be obtained. Very beautiful lace patterns are produced by wetting a warm glass plate with absolute alcohol, and dropping tallow, stearic acid, or spermaceti upon it ; all the finer portions of the film are at once dissolved, while the veins and denser por- tions remain, reminding one of skeleton leaves (see photograph No. 2, Spermaceti); the thin films of butterine, dripping, and similar substances will not stand this treatment, but are at once dissolved. § 183. Cohesion Figures. — Tomlinson [Phil. Mag., 1861 and 1862), some years ago, drew attention to the peculiar cohesion figures of various liquids and oils ; but the patterns of the solid fats, when melted and dropped on to warm water, do not appear to have received any consideration. The author finds, how- ever, that each solid fat behaves differently, and may also in this way be identified, and any admixture generally be correctly surmised. Should the water be at such a tempera- ture as to keep the fat very fluid, it rapidly spreads over the surface of the water, breaks up into lacunse, shows a beautiful iridescence, and the phenomena are over so rapidly as to leave but little impre.ssion on the memory. The author, therefore, prefers to operate at temperatures just sufiicient to keep the fat a little fluid, so that the action takes place in a slow, regular, and methodical manner. As an example, one experiment may be detailed. Filtered pure butter-fat, butter adulterated with 5, and , with 10, per cent, of lard, and lard itself, were all put in the same air-bath and brought to 55°-5. A large flat dish made chemically clean was filled with water of 44°, and a single drop of each of the four fats was dropped sim\iltaneously on the surface of the water, and their behaviour noted. The butter-drop immediately spread itself out into a thin film, became agitated by a rapid circular motion, and threw off" minute droplets of butter-fat. The motion gradually ceased, the drop extended itself, became irregular in outline, crenated at the edges, and then contraction took place. At this stage its appearance was that of an irregular squai'e, surrounded by small circles at distances from the central square and from each § 183.] ANALYSIS AND ADULTERATION OF BUTTER. 319 other of some three diameters. Both butter-drops containing 5 and 10 per cent, of lard respectively, flattened out with extreme slowness, were agitated by a gyratory motion, threw off no droplets of fat, and ultimately broke up with considerable slow- ness. It was noticed that the 5 per cent, drop was thinner and larger than the 10 per cent. Tlie drop of lard underwent no alteration, remaining circular and quiescent up to the moment of solidification. W. G. Crook* has experimented upon the solvent action of carbolic acid on butter as compared with other fats. 1 grm. of purified butter-fat is put in a test-tube and liquefied, '2h cc. of carbolic acid solution (10 acid, 1 water) are added, after which the mixture is shaken, and then put on one side for a little time. If the sample is pure butter, it wholly dissolves ; if beef-, mutton-, or pork-fat is present, the mixture will resolve itself into two solutions of different densities, with a clear line of demarcation. If beef-fat, the lower layer will occupy about 49*7 per cent, of the total volume; lard, 49*6; and mutton, 44'0. W. Lenz has also tried this process, and generally confirms the results obtained by Mr. Crook. f P. Casamajor (Chem. News^ xliv., 309, 310) has proposed a novel method of distinguishing oleo-margarine from genuine butter. Pure butter at 15° has tiie same specific gravity as alcohol of 53"7 per cent., specific gravity = '926, and oleo-margarine as alcohol of 59*2 per cent., specific gravity = "918. Any butter, therefore, which is adulterated with oleo-margarine will float in alcohol of 53-7 per cent. In alcohol, 56-8 per cent, (mean of 53*7 and 59*2), pure butter when melted sinks, oleo-margarine floats. In alcohol, 59*2 per cent., butter sinks, whether solid or fluid. The amount of adulteration is calculated as follows : — Determine the strength of alcohol, the same specific gravity as the sample, let it e.g. = 57 per cent., from it take 53-7 per cent., which is a constant, being the butter expressed, as it were, in equivalent pei-centage of alcohol, and multiply the remainder by the reciprocal of the difference between the strengths of alcohol of the same specific gravity as oleo-margarine and butter = (57 - 53'7) 0-18 = 5-94 per cent, of oleo-margarine. By far the most valuable simple test is the behaviour of butter fat with acetic acid ; this is known under the name of Valenta's * Analyst, 1879, 1111. + Zeitschrijt fur analyt. Cheviie, ISSO, 370. C. Husson {Compt. Bend., 85, 718) has proposed an ancient test depending on the different solvent properties of alcohol for margarine, &c. ; and F. Filsinger has a very- similar method (Pharm. Central. Halle, xix., 42). Both these tests, how- ever, are of little practical value among such a number of positive reactions. 350 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 183. test. 3 cc. of the melted fat are poured into a test tube, an equal volume of glacial acetic acid, specific gravity 1"0563, added, and the whole heated to complete solution. The contents of the tube are now allowed to cool spontaneously, stirring the whole time with a thermometer, and at the point of turbidity reading the temperature — the limits of the point of turbidity of genuine butter are 56° to 62° C, of margarine from 98° to 100° ; any sample, therefore, above 62° C. should be further examined. The acetic acid should, in all cases, be tried on a sample or samples of genuine butter-fat ; if this is done, no mistake will be made. Jean* relies rather upon the amount of acetic acid dissolved; he places about 8 cc. of fat in a graduated test-tube, 1 cm. in diameter, immersed in water of 5U° 0.; then removes the excess of fat by a pipette until the fat is exactly 3 cc. at 50°; then 3 cc. of glacial acetic acid, specific gravity 1-0565, are introduced, the acid being measured at 22° C. ; the contents are wai'med for a few minutes, the tube corked and well shaken. The tube is then placed in the water at 50°, and the volume of undissolved acetic acid read oft'. Nine samples of butter averaged 63-33 per cent, of acetic acid dissolved; margarines vary from 27° to 32°. It is obvious that the turbidity test, and the amount of acetic acid dissolved can be done on the same sample at the same time. M. Crismerf has proposed a turbidity method under the name of "the critical temperatui'e of dissolution," which diflers from the Valenta test, but is probably equal to it in value. About half a cc. of the filtered fat is introduced into a tube of small diameter; to this is added about 0-75 cc. of alcohol; the tube is hermetically sealed and attached by means of a platinum wire to the bulb ot a thermometer ; the bulb and tube are immersed in a small sulphuric acid bath, and the temperatui'e slowly raised until the meniscus separating the two layers becomes a horizontal plane. At this point the thermometer and tube are withdrawn from the bath, and the two liquids mixed together by shaking the tube with the thermometer ; they are then again placed in the bath and the temperature allowed to fall, the thermometer with attached tube being shaken all the time. The moment in which there is a marked turbidity is noted, and tliis is considered the temperature of dissolution. Fourteen genuine butters gave from 98° to 102°; mean 100°. Margarines varied from 122° to 126°. * Corps gras indu-ttriels, 1892, xix. 4. t Bull de I'Aasoc. Bebje des Chimistes, ix., 1895; also abstract in Analyst, September, 1S95. § 184.] ANALYSIS AND ADULTERATION OP BUTTER. 351 § 184. (6.) The Melting-Point. — Various methods have been proposed for the determination of the melting-points of fats. The one used by most analysts is to take the melting-point in a fine tube. A piece of quill-tubing is drawn out, so as to make a tube about the diameter of a knitting needle, and from 2 to 3 inches in length. The fat is now drawn up to the extent of about an inch, and permitted to solidify. The tube thus charged is placed in some cold water in a small beaker, which is " nested" in a second beaker, a little water being between the two, the inner beaker carrying also a thermometer. Heat is now applied, and the moment the fat runs up the tube the temperature is noted. A modification of this process * is to take a short capillary tube, blow a bulb on it, and while the bulb is still hot, plunge the open end into the melted fat ; let it run up a short distance, and then solidify the fat by the application of cold. To take an observation, the tubes are placed in water, so that the bulb is uppermost ; on melting, the fat runs up into the semi- vacuous bulb, and this rise is somewhat more easily observed than in the simpler process. Another method is the employment of a little bulb weighted with mercury, so as to weigh from 3 to 4 grms. ; the bulb rests on the surface of the fat in a test-tube, which is immersed in a beaker of water provided with a thermometer, and the moment the bulb sinks is noted. A modification of this is the employ- ment of a light float sunk to the bottom of the fat, and retained there until it is solid; on now applying heat, the float rises at a certain temperature, which is taken as the melting-point. These processes are not entirely satisfactory, and different observers obtain results which do not agree well. Reinhardt,t takes the melting-point of fats as follows : — The fat is drawn up when melted into a fine tube, 6, immersed in a beaker of water, c ; the tube, h, is attached as shown in the diagram (fig. 35) to a simple form of pressure ap- paratus, consisting of a stoppered cylinder, g, the caoutchouc stopper of which carries a graduated thistle-head funnel, f, and is connected with the tube carrying the fat by the short Wght-angled tube, a; it is, therefore, possible to put a water Fig. .35. * 0. Kellner, Zeitschrift fur anal. Cheniie, xx., 1. + Repert. anal. Chem.', 1S86. 352 FOODS : their composition and analysis. [§ 184. pressure on the fat by filling up the graduated stem of the funnel to any desired height. To make comparative observa- tions, the tube, b, must be each time filled to the same height, and immersed the same depth in the water, the pressure must also be the same. Heat should be applied very gradually, and the thermometer, t, graduated so as to allow of fifths being read. The end of the observation is when the first bubble of fat is forced out and rises to the surface of the water. Disc Method of taking Mdting-Points. — K method of taking melting-points, adopted by the American Association of Official Agricultural Chemists at their meeting in Chicago, 1893, is thus described :* — The apparatus for determining the melting-point consists of (1.) an accurate thermometer for reading easily tenths of a degree ; (2.) a cathetometer for reading the thermometer (this may be done with an eye-glass if held steadily and properly adjusted; (3.) a thermometer; (4.) a tall beaker-glass 35 cm. high and 10 cm. in diameter; (5.) a test-tube 30 cm. long and 3-5 cm. in diameter; (6.) a stand for supporting the apparatus; (7.) some method of stirring the water in the beaker ; for example, a blowing bulb of rubber, and a bent glass tube ex- tending to near the bottom of the beaker ; (8) a mixture of alcohol and water of the same specific gravity as the fat to be examined. The discs of the fat are prepared as follows : — The melted and filtered fat is allowed to fall from a dropping tube from a height of 15 to 20 cm. on a smooth piece of ice floating in water. The discs thus formed are from 1 to 1-5 cm. in diameter and weigh about -00 mgims. By pressing the ice under the water the discs are made to float on the surface, whence they iire easily removed with a steel spatula, which should be cooled in the ice water before using. The mixture of alcohol and water is prepared by boiling, in two separate vessels, distilled water and 95 per cent, alcohol for ten minutes to remove the gases which they may hold in solution. While still hot the water is poured into the test-tube already described until it is nearly half full. The test-tube is then nearly filled witli the hot alcohol It should be poured in gently tlown the side of the inclined tube to avoid too much mixing. If the tube is not filled until the water has cooled, the mixture will contain so many air bubbles as to be unfit for use. These bubbles will gather on the disc of fat as the temperature rises and finally force it to the top. * Chem. Keu-s, June 22, 1894. § 184.] ANALYSIS AND ADULTERATION OF BUTTER. 353 The test-tube containing the alcohol and water is placed in a tall beaker containing water and ice until cold. The disc of fat is then dropped into the tube from the spatula, and at once sinks until it reaches a part of the tube where the density of the alcoliol-water is exactly equivalent to its own. Here it remains at rest and free from the action of any force save that inherent in its own molecules. The delicate thermometer is placed in the test-tube, and lowered until the bulb is just above the disc. In order to secure an even temperature in all parts of the alcohol mixture in the vicinity of the disc, the thermometer is moved from time to time in a circularly pendulous manner. The disc having been placed in position, the water in the beaker-glass is slowly heated, and kept constantly stirred by means of the blowing apparatus already described. When the temperature of the alcohol-water mixture rises to about 6° below the melting-point, the disc of fat begins to shrivel, and gradually rolls up into an irregular mass. The thermometer is now lowered until the fat particle is even with the centre of the bulb. The bulb of the thermometer should be small, so as to indicate only the temperature of the mixture near the fat. A. gentle rotatory movement should be given to the thermometer bulb. The rise of temperature should be so regulated that the last two degrees of increment require about ten minutes. The mass of fat gradually approaches the form of a sphere, and when it is sensibly so, the reading of the thermometer is to be made. As soon as the temperature is taken, the test-tube is removed from the bath and placed again in the cooler. A second tube, containing alcohol and water, is at once placed in the bath. The test-tube (ice water having been used as the cooler) is of low enough temperature to cool the batli sufficiently. After the first determination, which should be only a trial, the temperature of the bath should be so regulated as to reach a maximum of about l°-5 above the melting-point of the fat under examination. The distilled water for floating the piece of ice on which the discs are made should be recently boiled, to free it of all air particles. The edge of the discs should not be allowed to touch the sides of the tube. This accident rarely happens, but in case it should take place, and the disc adhere to the sides of the tube, a new trial should be made. Triplicate determinations should be made, and the second and third results should show a near agreement. 24 354 FOODS : their composition and analysis. [§ 184. Example. — Melting-point of sample of butter .— - First trial, 33°-15 C. Second trial, 33°-05C. Third trial, 33°-00 C. The following melting-points are taken by the old methods, and are somewhat high, but are given as the values usually accepted : — Margarine, 31° Cocoa butter, 34° Butter (average), ..... 35° Beef-dripping, 43°' Veal-dripping, ...... 47° Mixed, 42° Lard, from 42° to 45° Ox-fat, from about 48° to 53° "0 Mutton-fat, from ..... 50° to 51°*6 Tallow, 53° -3 It hence follows that a low melting-point indicates the pro- bable presence of margarine, especially that from which is partly manufactured from a concrete oil, obtained from the seeds of Garcinia Indica, and is known under the name of Mangosteen oil, or kokum butter. A higher melting-point indicates, as a probable adulterant, dripping, lard, or other animal fat. The Titer Test. — A melting-point method, which gives useful results, is the "freezing" point of the fatty acids ; unfortunately it requires considerable material. At least 50 grms, of the butter-fat are saponified, the fatty acids separated and allowed to solidify ; the acids are melted and poured into a test-tube 25 cm. long and 16 cm. wide, filling the tube half full. Tlie tube is put into the neck of a suitable flask, and a delicate thermometer inserted ; when the mass begiiis to cloud, the thermometer is given a rotatory movement, and the mei'cury watched. At first it falls regularly, then stops, and gradually rises one or more tenths of a degree, to again stop, and then fall as before. The last stationary point is called the " titer " or solidifying point. It is advisable in all the above processes to allow the melted butter or fatty acids^ as the case may be, to " set " a definite time previous to the determination. The same fat melted and cooled will give two or more melting-points, if such melting- points are determined at different intervals of time, it is best then, in all cases, to prepare the fats one day and determine the melting-points on the following day. § 184.] ANALYSIS AND ADULTERATION OF BUTTER. 355 (c.) ApiMcation of the Refractometer to the Testing of Butter-fat. — J. Skalweit * has made some determinations of the angle of refraction of various fats, and believes there is a sufficient difference between the angle of butterine and pure butter to enable substitutions and adulterations to be detected ; he uses an Abbe's refractometer. The fat, kept for some time at 20°, is spread out on a watch-glass, and covered by a piece of Swedish filter-paper ; the fat is absorbed by the paper, and a clear gre;ise spot forms in the centime. The grease spot is applied to the edge of the Nicol prism to which it readily adheres; the apparatus is closed, and the angle estimated at 20°. The following results re given : — Water, . 1-333 Olein, from commercia oleic acid, . 1-46.35 Oleic acid, at 17°, 1-4638 Do., at '20°, . 1-46.39 Genuine Butter, 1-4652 Do., 1-4658 Cocoa Butter, . 1-46S0 Lard, 1-4690 Margarine, 1st quality, 1-4692 Do., 2nd „ 1-4720 Do., 3rd „ 1-4796 Do. Oil, 1-4680 Butterine, 1-4712 j 1-4698 i Hanoverian m 1-4698) Do., Do., Do., Do., 1-4733 English make. Refined Cotton-seed Oil, 1-4748 Crude „ „ 1-4732 Cod-liver Oil, . 1-4S01 T jnseed Oil, . 1-4835 The author lias made a number of determinations of angles of refraction by Abbe's insti'ument. The apparatus was enclosed in a copper air-bath, open at the top, but provided elsewhere with double walls, between which were placed glycerin and water; the boxing was completed by a cover of caoutchouc, in which a slit was cut to allow of the eyepiece protruding through; a sni;ill window in the copper box admitted light to the mirror, liy carefully heating this copper bath, the whole instrument could be maintained at any desired temperature. * Abbe's refractometer is fully described in "Neue Apparate zur Bestimmung des Brechungs-u. Zerstreuungs-vermogens fester u. fliissiuer Korper." Von Dr. E. Abbe. Jena, 1874. Tlie instrument is very conip.iot, and estimates both refraction and dispersion; the method of using iL is extremely simple; it can be obtained from Zeiss of Jena. 353 FOODS : their composition and analysis. [§ 184. It is neither advisable nor necessary to place the drops of fat on filter paper, it is best to examine a thin film between the prisms themselves, the edge of the shadow is then sharply defined. By means of the long arm, called by the inventor the " alhidade," the shadow is adjusted so that its border is exactly coincident with the cross lines, and the field is made colovii'less by working a milled head which regulates and measures the dispersion. These operations the author performs in the copper air-bath, itself, and if after 15 minutes interval at any desired temperature there is no further change in the refraction, the instrument is removed and the index read. To ensure accuracy a second observation is desirable, and the mean of the two taken. The best temperature for observation the author concludes to be the same one at which the specific gravity is taken, viz., 38°. At that temperature the most uniform results were obtained, and this temperature gives the necessary data for the calculation of what Dr. Gladstone has termed "the specific refractive energy," i.e., the angle of refraction, minus unity, divided by the density. In the following tables, p. 357, the angle of refx'action, the specific refractive energy, and various other data are given as determined from genuine and adulterated butter-fats. The angle of refraction in milk-fot, as extracted by ethei-, and genuine butters, varies from a maximum of 1*4575 to a minimum of 1*4543 at 38°; the mean is 1*4562, and the average specific refractive energy is •502. The refractive angle of margarine is always above 1*4620, generally 1*4639 or 1*4640, at 38". A mixture of 50 per cent, by weight of margarine (specific gravity •90578, melting-point 25°*5, refractive angle 1*4624) and 50 per cent, of butter (specific gravity -91073, melting-point 29°-8, refractive angle 1-4577) gave an angle of, at 38°, 1*4595, to which a genuine milk-fat may attain; the same materials, mixed in the proportion of 25 per cent, of margarine and 75 per cent, of butter, gave 1*4582 as the angle at 38°. Hence it is certain that to the analyst the refractometer has some utility, although it will not assist much in the case of moderate mixtures; but what may be said is that a butter at 38°, having an angle at or below 1*4570, is almost certain to be genuine, but above that angle is probably adulterated. § 1«4.] ANALYSIS AND ADULTERATION OF BUTTER. 357 TABLE XXa. — The Angle of Refraction, Specific Refraction, Melting-Point, Specific Gravity, and the Results of the Reichert and Koettstorfer Tests as represented in 33 Genuine Butters. Angle of Refraction at 38°. 1-4543 1-4565 1-4570 1-4556 l-4o70 1-4575 1-4562 1-4558 1-4575 1-4565 1-4565 1 4560 14570 1-4575 14570 14555 1-4565 1-4555 1-4555 14555 1-4535 1-4555 ] -4555 1-4555 1-4560 1-4568 1-4560 14560 1 -4562 1-4565 1-4560 1-4570 14555 Max. 1-4575 Min. 1-4543 Mean 1^4562 Specific Refraction (i.e., Angle of Refraction, - 1 divided by density). •50037 •50145 -51094 •50147 •50155 •50088 •49969 •60174 •511018 •50004 •49944 •50051 •50068 •60097 •61(171 •49962 •49918 •49887 •4990O •49937 •49S90 •49987 •49910 -4S987 -57236 -49570 •50023 •50055 •50012 •51140 •51270 •61061 •51270 •49887 •50212 -911-58 •91233 •91138 •91-.M6 •91134 -91217 •91079 ■91217 •91085 •91266 •91293 •91310 •91290 •91378 •91225 •91266 •91369 •91251 •91307 •91281 •91-206 •91303 •91124 •91264 •91222 •91235 •91276 •91156 •91138 91-277 •91378 •91079 •91234 3o°-5 as'^o 32° 30°-5 So'-O 3.5°^0 31° 34°-8 S2°5 32°-5 33°-4 3'2°-0 35''-0 30°-5 cc.'s of d. n. Alkali used it Reichert's Test.* 14-4 14-9 139 14-3 136 141 140 13^9 15-3 14-5 150 1.5-2 15-5 15 6 15-0 10-0 14-9 154 16-0 14-4 16-0 13-3 14-8 Koettstorfer's 2241 224-1 2-21-2 220S 223-7 221-2 2244 2-22-9 234-0 2314 234-4 233-3 229-6 234-9 2;7-5 2-333 2-27-2 229 6 231 '5 2231 2342 2344 235 2 2338 2267 2319 332-9 220-6 232 Examples of Low-class and Adulterated Butters. Angle of Refraction at 38". Specific Refraction. Specific Gravity, at Melting Point. cc.'s of d.n. Alkali used in Reicliert s Test. Koettstorfer's Mgrms. of KHO used. t 1-4577 t 1-4585 § 1-4580 1-46-24 f 1-4595 ** 1-4582 •50255 •50422 •50293 •51050 •50550 •50.343 •91073 -90932 •91066 •90.578 •90902 •91016 29°-8 29°-0 30°-0 26°-0 26° -5 27°-0 1-2 7-4 114 2-6 7-7 9 7 2-30-7 213-9 2268 199 -4 212 8 218 4 * To convert these numbers into Reichert-Meissl value they must be multiplied by 2. t A low-class butter. J An adulterated butter, for which vendor was prose^juted and flned. § A low-class butter, probably adulterated. || Ordinary butterine. *|f 50 per cent, butterine, No. 4, and 50 per cent, of low-cluss butter, No. 1. •* 25 per cent, butterine, No. 4, and 75 per cent of low-class butter, No. 1. 358 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 184. The Oleo-Refractonieter of MM. Amagat and Jean* — This is a more convenient and better instrument for observations on fats than Abbe's refractometer, the differences between butters and margarines being greater. The centre of this instrument (fig. 36) Fig. 36. — Amagat and Jean's Oleo-Refractometer. consists of a circular reservoir made of metal, B B, closed by two plates of glass in front of two lenses, E and E', the one lens belonging to the collimator tube, S, and the other to the tele- scope tube, S'. In the middle of this reservoir is placed a receptacle in the shape of a prism, its glass sides having an angle of 107°. In front of the field glass of the telescope is a photographic scale, H (shown enlai'ged at K), and there is an adjustable shutter placed vertically, so arranged as to divide the luminous field into two parts, the one bright, the other in shadow. The readings of the apparatus arc made from the edge of the shadow. If in both reservoirs is placed the same liquid at the same temperature, tlien the shadow occupies the zero of the scale; but if a different liquid be placed in the inner prism, there will be, according to its nature, a deviation either to the right or left, the amount of the deviation being appreciated by the numbers on the scale. The makers supf)ly a typical oil having no refraction, by means of which the instrument is set ; this oil is put in both receptacles, the temperature adjusted to 22°, and the shutter moved so as to mark zero. The oil is now run out of the prism, and replaced by the oil to be tested. Solid fats, such as butter or lard, are tested at a temperature of 45°. In all cases the oil or butter is to be freed fi'om free fatty acids. The butter is melted, filtered, dissolved in ether, and the ether shaken up with warm water ; the ethereal solution is thus freed from any soluble fatty acids ; it is then evaporated to dryness, and the dry washed fat tested. The rule is that vegetable oils and fats rotate to the right, animal oils and fats to the left, thus — * Ladan Bockairy in Cli. Girard and A. Duprt^'s Anali/se des Matieres Alimtntairea, Paris, 1894 ; also Muter in Analyst, May, 1S90. § 184.] ANALYSIS AND ADULTERATION OF BUTTER. 359 Olive oil, . . . + 1 -5 to 2 Colza ,, . . . + 16-5 to 17-5 Ground nut oil, . + 4-5 Sesame oil. . + 17 Cotton ,, . +20 Linseed ,, . +53 Castor ,, . +40 Heinpseed oil, . +33 Poppy „ . +30 Almond ,, . + 6 Japonica ,, . +50 Neatsfoot oil, , . . . — 3 Horsefoot ,, .... - 12 Lard, - 12-5 Beef tallow, .... - 16 Mutton „ .... - 20 T, , , f . f - 21 to - 36 ^'^"^'^-^^*'t average about -30 Margarine, .....— 15 Cocoa-nut oil deviates to the left like an animal fat ; it deviates some 36°. (d.) Specific Gravity. — One method of obtaining the specific gravity of butter-fat is to fill a counterpoised specific gravity bottle, provided with a thermometer stopper, of 50 to 100 grms. capacity, with water of 35° (95' Fahr.), and immerse it in a beaker of water of about 43° (109°-4 Fahr.). By thus heating the specific gravity bottle by a liquid which is falling in temperature, the water in it can be brought exactly to 37°-7 (100° Fahr.), at which temperature the bottle is taken out, slightly cooled and weighed; and in this manner the weight of that particular bulk of water at 37° "7 (100° Fahr.) is obtained, and this value used for the subsequent operations. To take the specific gravity of the fat, the pure filtered fat, at 35° (95° Fahr.), is poured into the clean dry bottle, and the exact process just detailed followed. Latterly many chemists have preferred to take the specific gravity at 100° C. Butter, as compared with water at 15°, has a gravity at 100° C. from 0-866 to 0-870; margarine, from 0-859 to"0-862. Skalweit has found that the greatest difi"erence in gravity is to be found at the temperature of 35°, and therefore recommends the gravity to be taken at that temperature ; thus, butter-fat at 100° 0. had a gravity of 0-8672, butterine 0-8598 — a diflerence of •0074 ; but at 35° the butter gravity was 0-9121, the butterine 0-9019— a diff"erence of -0102. Gravity at 100° may be taken either in a specific gravity bottle or by means of a Westphal's balance, but best of all by a Sprengel tube. In the former edition of this work Mr. Wigner's proposal to use specific gravity bubbles was mentioned ; in butters of 0-911 specific gravity (at 38°) a bead of specific gravity 0-8S9 slowly sinks at 63° C. ; but the process is not sufticiently accurate to have found favour among analysts. The specific gravity, as first pointed out by Mr. Bell, of Somerset House, has a direct relation or correspondence to the percentage of insoluble acids, a fact, it must be remembered, only applicable to pure unadulterated butter-fat. Thus — • Specific Gravity Actual Insoluble at37°-7 Acids Founa. (100° F.) Percent. •913S2 87-47 •91346 87-89 •91337 •91290 87-98 S8-48 * Muter, Analyst, Specific Gravity at ;i7°7 (100° F.) •912S6 •91-276 •91258 •91246 ., p. 7, 1877. Actual Insoluble Acids Found. Per cent. 88-52 88-62 88-80 89-00* 360 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 184. The fats used for the adulteration of butter are of low density. Vegetable butterine lias a specific gravity -90294, dripping -90659, so that a low specific gravity — that is, anything below -91101 — is strongly indicative of foreign fat. (e.) 21ie Viscometry of Butter. — Killing* has described an apparatus for taking the viscometry of butter; it essentially consists of a 50-cc. pipette, provided with a thermometer within the pipette, and a glass cylinder which acts as a jacket to the pipette and enables the contents to be maintained at a suitable temperature. The cylinder is also provided with a thermometer. The butter-fat, properly clarified, melted, and brought to a tem- perature of 40° -5, is sucked up into the pipette above the mark, then water at 42° is poured into the outer jacket, and the butter- fat, when both thermometers stand at 40°, run down to the mark, so as to measure exactly 50 cc. ; then the tap of the pipette (which is at the top, not at the bottom) is opened wide, and the time accurately noted which the butter takes to run into a beaker to a mark below the body of the pipette ; this is com- pared with the same quantity of water at 20° run in the same manner from the same cylinder, water being taken as 100 ; thus, if a butter-fat took 222 seconds and water 80 '33 seconds, the calculation is — 222^00^^00 _ 80-33 ~ "'^ '^ Various margarines gave a mean time of 4 minutes 12-77 seconds, equal to a viscosity number of 314*7; various butters, a meantime of 3 minutes 43-76 seconds; mixtures gave propor- tionate numbers. Lard and beef-fat both took a longer time than margarine. By carefully standardising a jiarticular pipette for water, for margarine, and for pure butter, this jjrocess may give useful results. R. Brullef distinguishes butter from margarine by treating 5 cc. of the clarified fat at 130° with a little pumice-stone and 8 drops of fuming nitric acid, the whole mixed and heated for twelve minutes; it is then cooled to 21°, and, after one hour, tested in an "oleo-grammeter," which consists of a rod gliding on a bearing, and surmounted by a little table on which weights can be placed. On an average, 250 grms. are required for pure butter, and 5,000 grms. for margarine, mixtures giving inter- mediate figures. Should this process be used, each analyst had better obtain his own standard. {/.) The Acid Value. — The acid value of a fat is the number of milligrammes of potassium hydrate required to saturate the free * Zeit.f. angewandle Chemie, 1894 and 1895. t Comptes Jiendus, 1893, cxvi., 1255. § 185.] ANALYSIS AND ADULTERATION OF BUTTER. 361 fatty acids in 1 grm. of the fat. This becomes of importance in the examination of rancid butters, but in butters whicli are not rancid it is seldom estimated. A weighed quantity of the melted, filtered, clear butter-fat is heated in a mixture of 2 parts of ether and 1 of alcohol, on the water-bath, until a perfect solu- tion is obtained; phenol-phthalein is added, and then the solution is exactly neutralised by decinormal alkali. Thus, 10 grms. of butter-fat, dissolved as above, required for neutralisation 3 cc. of decinormal potash ; since each cc. of d. n. potash is equal to 5-61 mgrms. of KOH, it follows that the 3 cc. are equal to 16-83 mgrms. of KOH; therefore, as 10 grms. were taken, the acid value is one-tenth of this — namely, i-683 mgrms. § 185. {g.) Direct Titration of Butter by Alcoholic Potash. — This elegant test was first suggested by Dr. Koettstorfer, and is a general method more or less applicable to all fats, but especially suited to butter-fat, for most other fats contain only the higher fatty acids ; as the lower acids have a smaller molecular weight, butter must contain more molecules of acid than equal weights of another fat — in other words, requires more potash for saponi- fication. Koettstorfer used semi-normal hydrochloric acid and an alcoholic solution of potash, this alkaline solution being about the same strength as the acid ; the indicator was a dilute alcoholic solution of phenol-phthalein. From 1 to 2 grms. of the purified filtered fat are weighed in a tall beaker of about 70 cc, capacity, 25 cc. of the potash solution are added, and heated in a water-bath. When the alcohol is nearly boiling, the mixture is stirred with a glass rod till all the fat is dissolved, which does not take more than a minute. The glass rod is washed with a little alcohol, and the beaker covered with a watch-glass, and heated further for fifteen minutes, in such a manner that the alcohol does not boil too violently. At the end of the quarter of an hour, the watch-glass is washed with spirit, and the alcoholic solution is stirred for one minute longer with the glass rod before used, so as to saponify any fat that may still adhere to it. The solution is now taken from the water-bath ; 1 cc. of an alcoholic solution of phenol-phthalein added, and it is titrated back with semi-normal hydrochloric acid. The exact point is very sharply indicated by the phenol-phthalein changing from a crimson to a yellow. (This reagent is very sensitive to COo ; it is therefore better to use a flask than an open beaker.) Thirteen butters treated in this way by Dr. Koettstorfer used for every grm. of fat from 221-5 to 232-4 mgrms. of KHO. Thirty-three genuine butters examined in the author's laboratory gave as a maximum 332-9 mgrms., as a minimum 220-6 ragrms» KHO (see Table at p. 357). 362 FOODS : their composition and analysis. [§ 186. On the other hand, there is a wide difference between this amount and that required by other fats, the following being about the saturation capacity in mgrms. of potash for 1 grm. ■of various fats : — Potash KHO. Milligrainmes. Oleo-margarine, ....... 195 "5 Beef-dripping, 196 '5 Tallow, 196-8 Lard from kidneys, ...... 195 "8 Lard from unsmoked bacon, . . . . . 196'7 Commercial lard, . . . . . . . 195'0 Dripping, 197-0 •Or if the suggestion of Mr. Allen* be accepted, and the results be translated into equivalents of the fat by dividing 56-1 by the mgrms. of potash, the results are as follows — Oleo-margarine, ....... 286-5 Beef-dripping 285-5 Tallow, 285-1 Lard from kidneys, ...... 286 -5 ,, ,, unsmoked bacon, . .... 286-7 Commercial lard, 287-1 Mutton drippiug, 284-8 The chief convenience in expressing the number in equivalents is, that it then becomes a matter of indifference whether potash or sodat is \ised for the saponification. The practical question in the use of this test is : what is the lowest limit above which a butter may be passed as genuine, but below which it will be necessary to examine the butter by other means 1 The general opinion of analysts as to this point is, that butter-fat, 1 grm. of which uses less than 226 mgrms. of KHO (equivalents 248-2), is probably adulterated. The formula for calculating the amount •of admixture which has been proposed is (•227 -n) X 3-n = X. X being the percentage of admixed fat, n the number of mgrms ■of potash used. § 186. {It.) The DecomjyositioQi of the Fat into Fatty Acids and Glycerin. — This is eftected by saponifying with an alcoholic solution of potash, decomposing the soap with sulphuric acid, washing the subsequent fatty acids with water, titrating the soluble, and weighing the insohible acids. The details of the process have been so simplified by successive improvements, that what was formerly a tedious and even difficult operation, is now * Analyst, 1879, 162. "I" It is scarcely necessary to add that should soda be used, then 40, the equivalent of soda, must be divided by mgrms. of the alkali used. 186.] ANALYSIS AND ADULTERATION OP BUTTER. 3G3 The solutions requisite are as moderately speedy and simjilt follows : — * (1.) Approximately semi-normal alcoholic potasli solution, 28 grms., roughly weiglied, of KHO, dissolved to a litre with alcohol (specific gravity -840). (2.) Approximately semi-normal sulphuric acid— i.e., 25 grms. of the strong acid to the litre. (3.) Deci-normal soda solution of exact strength, most con- veniently made by dissolving metallic sodium in water, in the exact proportion of 2 '3 grms. to the litre. [1 cc. equals -0088 of butyric acid.] It is necessary to know with the greatest exactitude the rela- tionship between the potasli and the sulphuric acid solution ; the exact quantity of alcoholic potash that is to be used in the analysis is delivered from a 25 cc. or 50 cc. pipette, as the case may be, phenol-phthalein solution added, and then titrated by the acid. It is also necessary to know the relationship between the d. n. soda and the sulphuric acid, which mvist be found in the iisual way. 4 to 5 grms. of the ]nire dry fat are weighed by difierence into a flask, and 50 cc. of potash solution added ; the flask, closed by a glass marble, is now heated on the top of the water oven, and by occasionally giving it a rotatory motion, saponi- fication is complete under the hour at the low temperature of 50° The author does not himself fol- low the above process, but uses the strong small assay flasks recom- mended by Dr. Dupre. These flasks are of about 70 cc. capacity, and with rather long narrow necks, the whole capable of bearing consider- able pressure. 4 to 5 grms. of the fat are poured into such a flask, 25 cc. of potash solution added, well corked with a caoutchouc stoi)per, which must be secui'ed by string and strong linen or canvas, and then the flask susjjended in the boiling water of a water-bath. At the end of an hour or less it may be taken out completely saponified.t When cool the flask is opened, the soap gently melted and poured into * " Butter-Fat," by E. W. Jones, F.C.S. Analyst, May, 1877. t The reason for preferring this method is, that less potash is required. Fig. 37. 364 FOODS : their composition and analysis, [§ 18G. a flask of about 500 cc. capacity, having a long, rather narrow- neck (see fig. 37), which carries the tubes a and h — the tube a for the admission of air, the tube h furnished with a stopcock. In this flask the soap is decomposed by adding about 1 cc. more sulphuric acid than is necessary to neutralise the potash; if, for example, the latter is neutralised by 25 cc. of the sulphuric acid 26 cc. are added, and after this addition the fatty acids melted so as to form a layer on the surflice of the acid water. At this point the whole may be diluted with warm water up to 200 cc. or 300 cc, the cork carrying the tubes inverted, and the flask turned upside down, as represented in the figure. After standing a few hours the cake is more or less solid, and the lower stratum of liquid may be run off" almost clear. It will, however, always be safest to pass it through a filter. By adapting an india-rubber tube to a, warm water may be sucked up through h, and the fat washed in the flask (perfectly closed by pinching the india- rubber), and then the cake allowed to form as before. The fluid is now again run oflf from the solid, and this time cold water may be sucked up through o, and the whole process of alternations of hot and cold water repeated. Lastly, the cork with its tubes is removed, any adherent fat washed oflf with warm water into the flask, the latter adapted either to an upright Liebig's condenser and boiled, or connected in the usual way with a Liebig, which has as a receiver a flask, adjusted by a cork tightly to the bent tube of the condenser, and is furnished with a mercury valve, the whole forming a closed system. In the latter case, also, the heat is applied to boiling for five or ten minutes, and the distillate added to the filtrates; lastk, the cork with tubes is again connected, the flask inverted, the liquid when cool run off, and the fat flnally washed with a little cold water and allowed to drain. The watery liquid contains sulphuric acid, glycerin, sulphate of potash, alcohol, butyric, and the other soluble fatty acids; it will be in bulk from 600 to 700 cc, and may be made up to any definite quantity. In any case, a portion of it — a quarter, a fifth, or even a tenth — must be taken and titrated witli d.n. soda, which, when the quantity required to neutralise the 1 cc. of sulphuric acid in excess is subtracted, indicates the amount of soluble acid, and is always returned as butyric, which is near enough to the truth. Instead of this method it may be useful to distil the acid liquid until all the volatile acid which can be obtained has gone over, and then titrate the distillate. It is also possible to separate the volatile fatty acids from such a solution by shaking up with ether in the tube figured at p. 69, the ether dissolving the acids freely. § 186.] ANALYSIS AND ADULTERATION OF BUTTER. 365 The insoluble fatty acids I'emaining partly in the flask, with a trace on the filter, are now united in a flat porcelain dish. This is done by melting the acids in the flask, pouiing off", and extract- ing by alcohol and ether — -the same solvent also dissolving the acids from the filter. On evaporation of the alcohol and ether, one or two large bubbles of water may be formed in the acids, and it is best to add a few drops of absolute alcoliol. The dish is now placed on the top of the water-bath (the water in which should only boil gently), and weighed at short intervals ; if after twenty minutes only 1 or 2 mgrms. are lost, the weight is con- sidered constant. The following are a few examples of percentages of fatty acids found in genuine butters : — (1.) (2.) (3.) (4.) Soluble, . . 5-92 576 5-37 477 Insoluble, . 87 -86 88-10 87-68 88-44 9378 93-86 93-05 93-21 It is generally accepted that 88 per cent, of insoluble acids, if associated with 6-3 of soluble acids, is a fair standard of butter calculation, and that if a butter shows anything less than 89-5 insoluble, with 5 soluble, it may be passed as genuine. A few examples of adulterated butter-fat are as follows : — (1.) A Commercial Butter. (2.) A Commercial Butter. (3.) Margarir Soluble, . Insoluble, • 1-98 . 93-o0 2-34 93 82 -58 95-51 95-28 96-16 96-09 (/.) Reichert-MeissVs Process. — 5 grms. of the melted clear fat are weighed in a flask of 200 cc. capacity, and saponified by 50 cc. of alcoholic potash (4 grms. of KHO in 100 grms. 70 percent, alcohol). The alcohol used must be pure, and free from aldehyde or acid. The soap solution is heated on the water-bath until it Tio longer smells of alcohol. The soap paste is dissolved in 100 cc. of water, 40 cc. of 10 per cent, sulphuric acid are added, with a few pieces of pumice-stone, and the flask is fitted to a Liebig condenser by means of a T-piece provided with a tube. The liquid is distilled carefully until, in about an hour's time, 110 cc. have passed over; 100 cc. of this are filtered through filter paper. To the filtered liquid phenol-phthalein or litmus is 366 FOODS : their composition and analysis. [§ 186. added, and the liquid is titrated with d. n. potasli. The number of cc. used is increased by one-tenth. English butters made from the mixed cream of several cows have a mean value of 29, and the minimum value adopted in this country is 24, but smaller values may be obtained from the butter of single cows, and under these circumstances the minimum may fall as low as 20-4. Eeichert and Meissl have proposed to calculate the amount of real butter-fat by taking 28-78 as a standard by the following form.ula : — 100 in - C) 28-78 - C n is the Reichert-Meissl value, and C the value of the admixed fat. Meissl thinks that C may be taken as 3. The formula is, of course, only a rough guide, because the value for the original butter may be higher or lower than 28-78. The following are results obtained by various observers with the Eeichert process : — D. N. Soda, cc. Butter-fat 24 to 32 Lard, -4 to -6 Rape oil, ........ "6 Sesame oil, ........ '7 Olive oil, '6 Palm oil I'O Kiduey-fat, "5 Cocoa-mit oil, 7-0 to 8 '0 Margarines, . . . . . . . . OS to 3*0 Butter-fat + 10 per cent, cocoa-nut oil, . . .26-8 50 „ „ ... 18-0 Butter-fat 50, cocoa-nut oil 22-5, margarine 27 'o, . 17-4 Eeichert' s process has been exhaustively examined by H. B. Cornwall & Shippen Wallace,* especially with reference to the question of minimum standard as determined by analysis of butters made from the milk of single cows. Their average results, based on the distillation of eighty pure butters made in three widely different parts of the State, are, that 2-5 grms., distilled as Eeichert recommends, give a distillate consuming 13 68 cc. of d. n. alkali. They found no connection between breed, food, or age of the cow, season of the year, or time after calving. The amount of volatile acid may consume as little as 11-3 cc. of d. n. soda for single cows, and they, therefore, think that it would not bfi safe to adopt a minimum standard above 11 cc. when 2-5 grms. of butter are taken, and it is known that the butter is from a single cow. It may here be remarked that the English analyst * C'he7n. Netos, Dec. 24, 1886. § 186.] ANALYSIS AND ADULTERATION OF BUTTER. 367 seldom has any history with the samples submitted to him, and hence but rarely knows whether the sample is a mixed one or not.* L. Medicus and S. Schererf have examined the process of Eeichert, and given a decided verdict in its favour. They have also examined by the process, whether the melting of butter is accompanied by any separation of its constitu.ents, and their experiments appear to prove that there is in point of fact a considei'able subsidence of the heavier fats. Thus, a kilogrm. of pure butter was allowed to cool slowly after melting in a capacious vessel, with constant stirring ; a sample was then taken, and the distillate, obtained after Reichert's method, used 28 cc. of d. n. soda. The same butter was then remelted, and allowed to cool slowly without stirring, and samples taken from vax'ious portions with the following results : — D. N. Soda. cc. Mixture, 28-0 Upper layer, 26 '6 Under layer 28 -J: Outer layer near the side, 28-8 Inner middle layer, . . . . . . 34'6 H. Kreist has proposed to saponify by sulphuric acid ; his: method has been subsequently modified by several chemists, the most important addition to the process being the use of potassic permanganate. Dr. Rideal adds 10 cc. of sulphuric acid, .specific gravity 1-836, to 2-5 grms. of melted butter-fat; the butter dis- solves. 100 cc. of water are now added, and the sulphurous acid developed is destroyed by about 1 cc. of strong permanganate solution. The mixture is distilled until 80 cc. of distillate are obtained. In nineteen different butters examined by the Reichert-Meissl process and the sulphuric acid process, the mean number of cc. of decinormal alkali consumed was 28-8 for the former and 29 '4 for the latter. A process of saponification has been proposed and practised by Mr. West Knight,§ which is based on the insolubility of the oleate, stearate, and palmitate of barium, and the ready solubility of the volatile fatty acid combinations with barium. The butter- fat is saponified with alcoholic potash in the ordinary way. The soap solution is diluted to 300 cc, and a solution of chloride of * Some results of the author's, with regard to the Reichert process, are to be found in the Table, ante, p. 357. Since these were examined by the original Reichert process (2 5 grms. ) the results must be doubled to compare with the above. + Zeitschrift Jur analyt. Chemw, 1880, j). 159. J Ohemiker Zeitung, 1892, 16, 1394. § Anolyd, IS80. 368 FOODS : their composition and analysis. [§ 186. barium added until a curdy precipitate separates, and the liquid is no longer rendei'ed milky by a fresh addition — the insoluble barium fatty acids are collected on a filter, and ultimately trans- ferred to a tube such as is used by Muter (p. 369), and the fatty acids liberated by sulphuric acid and shaken up with ether ; when separation has been effected, a fractional part of the ether is taken, and evaporated in a tared flask. Wm. Johnstone* has proposed a method of dealing with butter-fat which has the advantages of Koettstorfer's process, and gives the amount of soluble and insoluble fatty acids. 2-5 grms. of the butter-fat are saponified with a known quantity of alcoholic potash in a closed flask, the titre of the potash having been carefully ascertained. After saponification the liquid is exactly neutralised, and thus — A. The amount of alkali required to saturate all the acids of the butter is ascertained. The alcohol is now boiled off and an excess of acid added to •decompose the soap. The fatty acids are washed with hot water (as before described), the insoluble fatty acids collected on a filter and dried in air; the filter is transferred to a Soxhlet, and its contents exhausted witli dry ether, the ethereal solution being received in a weighed flask. When the extraction is complete the ether is driven off, the flask and its contents care- fully dried and weighed ; this gives — B. Insoluble acids by weight. Next the fatty acids remaining in the flask are saponified by a known volume of standard alkali and titrated with standard acid ; this gives — C. Amount of alkali required to saturate the insoluble acids. By subtracting C from A, obviously by difterence, the amount of alkali required to saturate D, the soluble fatty acids, is obtained. An example will make the working clear. 2-5 grms. of butter were saponified by 25-00 cc. of normal alcoholic potash ; after saponification the liquid was neutralised by 14-78 cc. of standard acid, therefore (A) the alkali was equal to 25-00 - 14-78 cc. ; that is, 10-22 cc. of normal potash. (B) The insoluble fatty acids weighed 2-2487 grms. = 89-95 per cent. The insoluble fatty acids were saponified by normal potash, the difterence of acid taken was equal to 8-13 cc. of potash — that is to say, the insoluble acids were neutralised by (C) 8*13 cc. of normal potash; but the total acids consumed 10-2 cc. of normal potash ; therefore, the soluble acids (D) must be equal to 8-13 - 10-22 cc, or 2-09, equal to 7-35 per cent, of butyric acid. Further Analysis of the Insoluble Fatty Acids. — The insoluble fatty acids are, as already stated, oleic, palmitic, and stearic ; it * The Analyst, xiv., 1SS9, No. 15S. §186. ANALYSIS AND ADULTERATION OF BUTTER. 369 is their total weight which is alone valuable, and to separate the three with accuracy is not easily effected. The first can, however, loe isolated by the following process, the details of which have been worked out by Dr. Muter. The process depends upon the well-known fact that the oleate of lead, Pb2CjgH330o, can be separated from plumbic palmitate, Pb2C\,3H3^0o, and plumbic stearate, Pb2CjsH350o, by taking advantage of the solubility of the former in ether. Muter's method of estimating oleic acid is as follows : — 3 grms. of the fat are saponified by means of alcoholic potash. The potash is carefully neutralised by acetic acid, using phenol-phthalein as an indicator. 200 cc. of water, to which 30 cc. of a 10 per cent, solution of plumbic acetate have been added, are boiled and, while boiling, the soap solution is slowly poured in with constant stirring. The whole is allowed to cool, the supernatant fluid poured off, and the lead soap washed with hot water by decantation. The precipitate is transferred to a stoppered bottle, 80 cc. of ether added, and, finally, the ether made up to 120 cc. The bottle is allowed to stand, with occasional shaking, for twelve hours, during which time the whole of the oleate will have dissolved. The ethereal solution is now filtered into a special tube (Fig. 38), the plumbic stearate, &c., being washed with ether until free from lead ; this usually entails the use of about 100 cc. of ether. Hydrochloric acid, 25 per cent, strength, is poured in up to the first mark on the tube and the contents shaken. The liquids are now allowed to separate and the acid layer drawn off by means of the stop-cock. Water is then poured in and the contents again shaken, the water being removed as before, and the whole process repeated until the washings are free from acidity. The total volume of the ether is noted, a fractional portion run off into an Erlenmeyer's flask, and the ether distilled until a very little ether remains. It is not distilled completely, so as to avoid exposing the oleic acid to the air; 50 cc. of pure alcohol are added to this residue and the solution titrated by decinormal soda, using phenol- phthalein as an indicator ; each cc. of d. n. acid is equal to 0-0282 oleic acid. Having thus ascertained the strength of the remaining ethereal solution in the tube, the next step is to run off as many cc. as correspond to 0*5 grm. of oleic acid into a stoppered bottle 25 Fi ¥ Fi2. 3S. Mutei'.s olciii tube. 370 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 186. of at least 350 cc. capacity ; this flask is also provided with a doubly perforated cork carrying two tubes, one of which is con- nected with a carbon dioxide apparatus. The flask is placed in warm water and the gas passed through until all trace of ether has been removed. To the residue 50 cc. of Hubl (see p. 372) are added, the stopper inserted and allowed to stand the usual time and then titrated.* E. Twitchell f has proved that 100 cc. of ether dissolves 0-015 of the lead salts of purified commercial stearic acid at 0°, and that this is an objection to Muter's process. He proposes a modification of Jean's process as follows, of which he gives the following example : — A lard, the liquid fatty acid of which gave when tested by Muter's process an iodine absorption of 94'1 per cent., was treated as follows : — 4 grms. of the fatty acids were dissolved in 50 cc. of 95 per cent, alcohol and 2-5 grms. of lead acetate in 20 cc. of the same alcohol added, both solutions being hot. The liquid with precipitate was put on one side for two houi's, the temperature being strictly 15° for the last hour. A part of the whole was then filtered into a separating funnel, treated with ether and HCl, the acids washed and dried in a current of COo, their iodine number obtained, and their percentage in the original solution determined. The precipitate was washed with 95 per cent, alcohol, decomposed with HCl, and the solid fatty acids dried and weighed; the results ■ Per cent. Iodine number. Solid fatty acids, 46-24 .... 4-9 Liquid fatty acids, 5182 . . . . 10.3-37 figures indicating that vW the solid acids and part of the liquid had been precipitated. If the iodine number of the total fatty acids is obtained the method of calculating results is as in the following example :— 4 grms. of fatty acids gave an iodine number of 62-57 ; the fatty acids from the filtrate at 15° gave an iodine number of 109-35 and were equal to 46-81 per cent.. The percentage of liquid acids multiplied by the iodine number of these acids and deducted from the iodine number of the total acids equals oleic acid. Dividing this by 0-9 gives the percentage of oleic acid precipitated with the solid acids. Thus in the above case ^^^^^^^^^^^ = 51 -19 - 62-57 = 11-38. Divid- ing 11-38 by 0-9 = 12-64 in the precipitate, which, added to 46-81, represents the total liquid acids, 59-45. * Analyst, April, 1889. fJourn. Amer. Chem. Soc, 1895, xviii., 289 to 295; Analyst, July, 1895. § ISG.l ANALYSIS AXD ADULTERATION OF BUTTER. 371 Mr. AVanklyn'^ has proposed estimating butyric acid derived from the formation of butyric etlier in saponifying with alcoholic potash as follows : — The butter is claritied in tlie usual way, and then 5 grms. are weighed and taken for tlie analysis. The butter is placed in a small retort of about 200 cc. capacity, and fitted to a condenser. About 100 cc. of alcohol (specific gravity 0-838) is added to the butter in the retort, and then 0-5 grm. of solid potish is added. The retort is then gently heated, and the contents are distilled, the distillation being continued to dryness. Tlie distillate is received in a bottle fitted with a stopper, and containing 40 cc. of accurately measured normal caustic potash or soda. When the distillation is complete, the stopper is placed in the bottle and the contents are shaken for a short time, and presently it will be found that the smell of butyric ether has vanished. Phenol-phthalein is now added to . serve as an indicator, and the solution is titrated with normal sulphuric acid. Good butters treated in this way yield from 2-8 to 34 per cent, of butyric acid as ether, while common butterine yields no trace. Estimation of Glycerin. — The determination of glycerin in fats, hitherto most unsatisfactory, can now, thanks to the labours of Wanklyn, Fox, Benedikt, and Zsigmondy, be accomplished with fair accuracy. . The process is based upon the oxidation of the glycerin by alkaline permanganate, the consequent formation of oxalic acid, carbon dioxide, and water; and the estimation of the oxalic acid; whence the glycerin is calculated according to the formula — CsHgOs + 3O2 = C2H0O4 + CO2 + 3H2O 100 parts of oxalic acid are equal to 102-2 parts of glycerin, or, since the oxalic acid is usually determined as lime carbonate, 100 parts of lime carbonate equal 92 parts of glycerin. The details of the process are as follows : — 10 grms. of the filtered butter-fat are saponified by a known volume of a solution of KtIO in pure methyl alcohol,! or, better still, by a strong aqueous solution of potash. Mr. Allen has recommended sapo- nifying the fat by means of 25 cc. of a 16 per cent, potash solution in a flask well stoppei-ed down by means of an india- rubber stopper, heat being applied by means of a water-bath for * " The Analysis of Butter," by W. Fox and J. A. \Yanklj-n. Analyst, 1SS4. t Ethyl alcohol cannot be used, for on treatment with boiling alkaline permanganate, some oxalic acid is produced. 372 foods; their composition and analysis. [§186. several hours, until from the homogeneous appearance of the liquid it is certain that saponification has been effected. The soap is now decomposed in the usual way by means of dilute sulphuric acid ; the fatty acids separated ; the filtrate made up to a known bulk, and a fractional part of tliis taken for oxida- tion. The oxidation is carried out as follows : — The solution is alkalised by potash solution contaitiiu'^ at least 5 per cent, of free alkali, and then a 5 per cent, solution of potassic perman- ganate is added until the liquid is blackish in colour. The solution is boiled, whereupon manganese oxide is precipitated, but the excess still tinires the liquid i-ed, the red colour is dis- charged by adding sulphurous acid, the whole is then filtered and the precipitate well washed with boiling water. The filtrate is boiled, and, while boiling, an excess of calcium acetate is added. The precipitate, consisting of an impure calcic oxalate, is collected, washed, dissolved in dilute sul- phuric acid, and titrated with permanganate solution, or it may be ignited, and the resulting calcic carbonate dissolved in d. n. HCl, titrating back with a known volume of a solution of d. n. soda, and using orange methyl as an indicator. The best results are obtained when the amount of glycerin to be oxidised is from -3 to -5 grm., and the dilution not greater than -1 per The Iodine Value.— Ku\>\ {Dlngl. polyt. Journ., 253, 281-295) has proposed a method which is of the greatest value in the examination of oils and fats generally, and can be applied to the analysis of butter. The melted fat is treated with an alcoholic solution of iodine in presence of an alcoholic solution of mercury bichloride ; under these circumstances the unsaturated ia,ity acids or their glycerides absorb iodine in a regular and definite manner. "With regard to butter, the only unsaturated fatty acid is oleic acid ; hence the amount of iodine absorbed has relation to the content of olein or to oleic acid. The following solutions are required : — * See papers by Messrs. Fox & Wanklyn, Chem. News, Jan. S, ISSG; by R. Benedikt & R. Zsigmondy, Analyst, x., 205; and by Alien, op. cit., xL, 52. J. David, Conipt. Rend. 94, 1477-1479, estimates glycerin as follows: — 100 grms. of the fat are melted, 65 grms. of barium hydrate are added, ■with brisk stirring ; when most of the water has been expelled the heating is discontinued; 80 cc. of alcohol of 95° are poured on the mass, and the whole stirred ; 1 litre of water is then added, and the mixture boiled for an hour. The barium soap remains insoluble, whilst the glycerol is dissolved by the water, which is freed from the excess of barium, reduced in volume by boiling, and finally evaporated in a vacuum at a low temperature. The glycerin might evidently be estimated by the oxalic method. § 186.] ANALYSIS AND ADULTEUATION OF BUTTER. 373 (1) Solution of iodine and mercury bichloride. 25 grms. of iodine are dissolved in half a litre of alcohol of 95 per cent, strength, and 30 grms. of mercury bichloride are dissolved in another half litre of alcohol ; the two solutions are now mixed, and allowed to stand, before standardising, for twenty-four hours. (2) Solution of sodium hyposulphite, 24 grms. to the litre. (3) 3-8747 grms, of potassium bichromate dissolved in a litre of water. (4) A solution of potassium iodide, 10 per cent, strength. The thiosulphate is standardised as follows : — 10 cc. of the potassium iodide solution are placed in a stoppered bottle, and 5 cc. of HCl added, together with 20 cc. of the bichromate solution. This will liberate exactly 0-2 grm. of iodine. The thiosulphate solution is run in carefully until a light straw colour only remains, then a little freshly-prepared starch solution is added, and the thiosulphate run in until the blue colour disappears. The num- ber of cubic centimetres added will, of course, be equal to 0*2 grm. of iodine. The next thing is to titrate with the tliiosul- phate the solution of iodine and mercury bichloride. It is best to take for this purpose 25 cc. of the iodine solution, and to operate in a similar way — that is, running in the thiosulphate until there is only a pale straw colour, and finishing with starch as an indicator. To obtain the iodine value of butter-fat, from 0-6 to 0-8 grm. of the clear melted fat is dropped on to the bottom of a tared flask, the flask and its contents weighed. The fat is dissolved in 10 cc. of chloroform, and 25 cc. of the iodine solution run into the flask, whicli should be at least 500 cc. in capacity; the flask is now- stoppered and put in the dark for four hours. Should the iodine solution become decidedly pale at the end of two liours, a second 25 cc. of iodine is run in, for unless an excess is present, accurate results will not be obtained. At the end of the stated time, 20 cc. of the potassium iodide solution are added (or, should there be a red precipitate of mercury iodide, even more), the liquid diluted with from 300 to 500 cc. of water, well shaken and titrated with the thiosulphate, using as an indicator starch solu- tion. The diflerence in the number of cubic centimetres of thiosulphate used on the original iodine solution, and that on the solution which has acted on the fat, gives the requisite data from Avhich to calculate the amount of iodine the fat has absorbed ; this is calculated into per cent, of the fat. Rowland Williams* has examined in this way thirty butters, the mean of which gave 35-34 per cent, of iodine absorbed, the extremes being 23-6 and 40-3. Seven margarines gave from 62-29 to 75-22 per cent. * Anahjst, June, 1S94. 374 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 186. The same process is applicable to the insoluble fatty acids, a weighed quantity from 0-2 to 0-4 of tlie fatty acids being dis- solved in chloroform and treated as above. The following are a few iodine values for various fats and oils : — Iodine number. Linseed Oil, ISS'O Castor Oil, 84-4 Olive Oil, 82-8 Lard, 59-0 Butterine, 55 -.3 Palm Oil, 51-5 Tallow, 40-0 ('ocoa Butter, . . . . . . . 34 '0 Nutmeg Butter 31-0 Butter-lat (from 2G-0 to 35-1), .... 31-0 Cocoa-nut Oil, ....... 8 "9 Japanese Wax, 4 2 In fats, like butter-fat, in which it is believed there are no other unsaturated acids than oleic, olein and the oleic acid can be in this way estimated. The theoretical amount of iodine absorbed by oleic acid is 90-07 per cent., and the theoretical iodine value for olein is 86"20. Hence, if the iodine value of a fat equals 1, the percentage of olein will be found from the following formula — 100 X 1 Olein = --ygvo- Or, simpler still, the iodine value multiplied by the factor 1-lGOl gives the olein. Similarly the amount of oleic acid in the fatty acids is found by multiplying the iodine value found by the factor 1-1102. Hehner* has proposed to treat fats with bromine, and weigh the product. .1 to 3 grms. of the fat are introduced into a wide- mouthed flask, and dissolved in a few cubic centimetres of chloroform ; then pure bromine is added drop by drop until in excess. The tlask is then heated in the water bath, and the bromine driven off by adding from time to time a little chloro- form. Finally the brominised fat is dried at 125°, and weighed. To compare the results with those obtained with iodine, the 127 gain in weight is multiplied by -— r- — that is, 1-587; the results fairly agree. Tlie Cryoscopic Method of examining Butter-Fat. — Raoult's method of determining molecular weights is applicable to butter-fat, and may be vised as a means of distinguishing it from margarine. F. Garelli and L. Carcano, using benzene as a solvent, and considering the constant for benzene to be 53 (it is usually taken as 49), found that pure butter had a mole- * Analyst, xx., p. 50. S 186. ANALYSIS AND ADULTERATION OP BUTTER. cular weis;l)t of from 696 to 716, while margarines ranged from 780 to 883.* The apparatus required is (1.) a Beckman's thermometer, D (fig. 39) ; this thermometer has a mercury reservoir at the top, e, so that variable amounts of mercury may be introduced into the lower bulb ; it has a range of only 5 or 6 degrees divided into tenths ; (2.) a tube with a side limb, A A'; and (3.) a wide test-tube into which the first one fits, so as to (as it were) jacket it with air. The whole is supported in a wide beaker, in which the freezing mixture is placed, and suitable stirrers, C and r, provided. The first thing is to ascertain the freezing point of a known weight of the solvent. The tube with side limb is weighed on the balance, and then about 20 grms. of the solvent are introduced, and the tube with its contents again weighed ; the weight subtracted from the tare gives the amount of liquid. Some ice and salt and water are placed in the beaker, and the whole apparatus arranged as in the diagram. The mercury of the thermometer gra- dually falls, and as it falls both stirrers must be worked ; on approaching the freezing point the thermometer becomes for a moment stationary, then suddenly rises, and then becomes stationary to again fall; when it last becomes station- ary is considered the freezing point of the liquid; the tube is taken out, the liquid allowed to melt, and the operation re- peated once or twice, the mean of the determinations being taken. Next from *5 to -8 grm. of the butter, or other fat. Fig. 39. Cryoscopic apparatus. is dissolved in the same liquid, and the operation again repeated ; this time there will be a difference of one or more tenths of a degree in the freezing poiut ; it will be lower than when operating on the pure substance ; this lowering is the molecular depression. The molecular weight is calculated by means of a constant, which, for the chief solvents, has already been ascertained. The two solvents ajiplicable for butter are benzene, for which the constant is usually taken as 49, and paraxylene,t which may be taken as * Analyst, March, 1894. i"Paterno and Moatemartini, Gazella Chimica Italiana, xxiv., ii., p. 197. 376 foods: their composition and analysis. [§186. 43. ^ Of the two solvents, paraxylene is much to be preferred, for it has the low freezing point of about 1 6° and the high boiling point of 13G°; hence, although expensive, it is readily recovered with but little loss. The author has made several determinations of the cryoscopic value of butter-fat ; but with paraxylene has obtained much lower values than Garelli and Carano. He is inclined to put the average molecular weight of butter-fat at about 580. An actual example of a cryoscopic determination with para- xylene will illustrate the above remarks : — Weight of paraxylene taken, ... 19-505 grms. Weight of butter-fat dissolved in the above, . r0784 ,. Depression of freezing poiut, .... 0-41°. If M equals the molecular weight, C the constant, G the per cent, of substance dissolved, and ;; the depression, then the mole- cular weight, M, is found from the equation — In the above example, replacing the letters by the experi- mental values — 5-528 x 43 :4^- = 579 -S Summary.- — In any case in which the adulterant is moderate in quantity,* up to 15 per cent, for exami^le, the analyst will have to exert all his skill, and analyse the butter-fat very thoroughly — and the methods of taking the melting-point, the amount of potash used u\) by a gramme of the fat, the specific gravity, the refrac- tion angle, the soluble and insoluble acids, the Reichert distil- lation process, the iodine absorption may all have to be used, for small percentages of mixtures can only be detected by carefully- considering a number of indications. Single tests may be used for sorting butters, but are not to be relied upon. It is fairly easy to make mixtures of certain fats which will yield the proper quantity of insoluble acid, and which would pass the iodine test, the specitic gravity, and the Koettstorfer tests, but not at all easy to make a mixture which will pass also the Reichert. LEGAL CASE. Somerset House standard for water in butter. At the Eath Police Court (January, 1879), a dairyman had been summoned for selling butter, the jiroximate analysis of which showed a considerable addition of water. An appeal to Somerset House elicited the following certificate: — " VVe hereby certify that we have analysed the butter, and declare the results of our analysis to be as follows : — * The false butters used to be for the most part either all margarine or heavy percentages of margarine, but now there are many mixtures in com- merce containing only small proportions of margarine. § 186.] BIBLIOGRAPHY. 377 Per cent. Water, •23-27 Butter-fat, 74-69 Curd, 1-26 Salt, -78 " The results of our analyses of numerous samples of ordinary commercial butters obtained from different parts of the country, iucludiug the south of England, show tliat the proportion of water present is very variable, and that it occasionally amounts to as much as 19 per cent. ' Water in Irish butter. A number of prosecutions for water in Irish butter were heard at the Manchester Police Court in 1894, before the Stipendiary, Francis John Headlam, Esq., the cases occupying seven days. Most of the leading analysts gave evidence. The cases were decided variously, according ta the circumstances of a warranty or otherwise. In two cases in which there was a discrepancy in the analysis by two different analysts, the magistrate stated that he would have convicted on 21 per cent, of water; the amounts of water on which the prosecutions were based varied from 214 to 2Q^ per cent. The Corporation of Manchester published the shorthand notes of the proceedings. BIBLIOGRAPHY. Angell, G., and Hehner, 0. — "Butter: its Analysis and Detection." Isted., Lond., Ib74; 2nd ed., 1878. Benedikt, 11. — Chemical Analysis of Oils, Fats, and Waxes. Translated and enlarged by J, Lewkowitsch, F.T.C. Lond., 1895. Besaira. — Sui metodi a distmguere il burro artificiale dal burro naturale. Lodi, 1888. Blyth, A. Wyntek.— Easy and Rapid Method of Manipulating the Fatty Acids. Analyst, ii., 1878. On the Figures or Patterns which Liquid Fats assume under certain Conditions. Analyst, 1880. Blyth, A. Wynter, and Robertson. — The Electrolysis of Butter-Fat. Journ. Ghem. Soc, 1889 (Proc), 5. Bockairy, Ladan. — "Beurre" in Girard and Dupre's "Analyse des matieres alimentaires. " Paris, 1894. Crook, W. G. — A New Method of distinguishing Butter from some other Fats. Analyst, 1879. DiETZEL, u. M. G. Kriesner.— Zei7sc?ir(/< /. an. Chem., 1879. DuPRE, A.— The Composition and Analysis of Butter-Fat. Analyst, 1, 87, 114. Foods and Food Adulterants. U.S. Department of Agriculture. Bull, No. 1.3, 1887. Faber, H. — On Water in Danish Butter. London, 1895. Fox, W., and J. A. Wanklyn. — Butter Analysis. Analyst, 1SS3, 73. Hager, H. — Pharm. Central. Halle, xviii., 413. Jean. — Chimie analytique des matieres grasses. Paris, 1892. Jones, E. W. T. — Butter-Fat: its Analysis and Composition. Analyst, ii., 1878, 19, 87; Influence from the Decomposition of Butters through Age on the Specific Gravity of the Fat, &c. Analyst, 1879. KoETTSTORFER, J. — Neue Methode zur Untersuchung der Butterfett auf fremde Fette. Zeitschrijt f. anal. CAem., 1879, 199, 431. Analyst, 1879, p. 106. Kretschmar, M.—Deut. Chem. Ges. Ber., x., 2091-2095. Medictjs, L., u. S. Scherer. — Zeitschrift f. avai. Chem., 18S0, 159. Milne, J. M. — Notes on the Analysis of Butter. Analyst, 1879, 40. 378 foods: their composition and analysis. [§ 187. Muter, J. — Butter Analysis. Analyst, 1877, p. 7. Perkins, F. V.—Anabist, 1879, 14-.'. Redwood.— On the Determination of the Melting-points of Butter and other Fats. Analyst, 1, 1877. E-EICHERT, E. — Butter-Priifung nach Hehner's Princip. Zeitschrift f. an. Chem., 18, 64. Reinhardt, C. — Ueber die Bestimmuns: des Schmelz-punktes der Fette. Zeit. f. an. Chem., Bd., xxv., 188'6. Skalweit, J.— Butter-Testiug. Analyst, 1886, 90. West-Knight, E. — On the Estimation of the Insoluble Fatty Acids in Butter-Fat. Analyst, 1880, 155. Wigner, G. \V. — On Estimation of the Gravities of Fats. Analyst, 1, 1877. 35. On Butter Analysis by Koettstorfer's Process. Analyst, 1879, 132. On Butter-Fat, and Ratio of Expansion by Heat. Analyst, 1879, 183. Wiley, H. — Butter and its Adulterations. tJ.S. Dep. of Agriculture — Chem. division. Bull., No. 13, 1887. WoLL. — Beitriige zur Butter Analyse. Zeit. f. an. Chem., 1887,28. Wright, C. II. Alder. — Animal and Vegetable Oils and Fats, Butter, and Waxes. Lond. 1894. Zune. — Traite general d'analyse des beurres. 2 vols. Paris et Bruxelles, 1892. BQTTERMILK. § 187. Buttermilk is the thin whey left behind when the fat has been extracted in the process of butter-making. It is never fat-free; it contains all the constituents of milk, but a great portion of tlie sugar has been changed into lactic acid. It is then essentially a dilute, poor acid milk. The average composition of fresh buttermilk is : — Water, 90-62 Casein, 3*78 Fat, 1-25 Milk-sugar, 3 38 Lactic acid, "32 Ash, -65 The lactic acid tends ever to increase, so that samples which have stood a little time will contain more lactic acid than the proportion above given ; besides the ordinary salts of milk it frequently contains the common salt added to preserve the butter. It is not an article of commei'ce, and from its occurrence merely as a bye-product differs in composition considerably.* * Some analyses by Dr. Vieth [Analyst, 1884) may be thus summarised: — Total Solids, Fat, Solifls, not Fat, Asli, per cent. per cent. per cent. per cent. Maximum, . . 10-70 2 51 10-lG 1-32 Minimum, . . 8-13 '49 7-13 -64 Mean, ... 9-43 '7 8-56 '75 §§ 188, 189.] 379 CHEESE. § 188. Cheese consists essentially of the coagulated albuminous matters of milk, especially of casein, with a variable quantity of fat, common salt, and alkaline and earthy phosphates. Cheese may be made from the milk of any animal, but the great majority of cheeses in commerce are made from that of the cow. The principle of the manufacture of cheese from fresh milk is very simple : the casein is precipitated by rennet, and carries down with it most of the milk-fat, as well as some portion of the milk-sugar. The thin whey is allowed to run off, and the precipitated "curds"* submitted to pressure, which has the effect of not only getting rid of the whey, but also of giving to the mass shape and consistency. Cheese may be made from sour milk without the addition of rennet, the lactic acid precipitating the casein ; but most of the cheeses in com- merce are made from fresh milk. Cheeses may be divided into two varieties — the soft and the hard ; the former are manufactured by precipitating with rennet at a low tempera- ture, and using but little pressure ; they have mostly an alkaline reaction. The hard cheeses are subjected to a higher temperature and stronger pressure, and have, when first made, an acid reaction. § 189. Soft Cheeses. — Examples of soft cheeses are cream cheese. Neufchatel (a Swiss cream cheese), /romac^e de Brie, and Cam ember t. Hard Cheeses. — Examples of hard cheeses are American, Cheddar, Stilton, Dunlop, Gloucester, and others. The genei-al composition of the chief cheeses of commerce may be gathered from the following table : — An analysis of curds by M. Rubner is as follows : — Water, Casein, Fat, . Ash, . Milk-sugar, Total solids, "Solids not fat, 39-73 32-40 Per cent. 60-27 24-84 7-33 4-02 3-54 380 FOODS : THEIR COMPOSITION AND ANALYSIS. 189. TABLE XX6.— Composition of Cheese. 1 J3 ^ 1 Analysts. 1 < ^ z American cheese, mean ) of six samples, . . \ 27-5 4-1 32-5 4-54 ( Chattaway, Pearmain, 1 and Moor. American clieese, mean \ of two samples, . . J 27-2 4-1 32-1 5-80 A. VVynter Blyth. Camembert, mean of \ two samples, . . . J 45 6 4-25 32 '2 3-63 / Chattaway, Pearmain, \ and Moor. Cheddar(English)mean "1 of three samples, . . J 34-2 4-10 28-5 4-30 Do., do. Cheddar (English) mean \ of two samples, . . / 28-1 3-34 22-5 7-3 A. ^Vynter Blyth. Cheddar (Canadian), . 33-3 3-60 30-6 4-3 \ Chattaway, Pearmain, \ and Moor. Cheshire, mean of two \ samples, .... J Cheshire, mean of six"! samples, .... J Cream (double), . . . 34-7 4-30 33-3 41 Do. , do. 44-6 4-61 30-7 4-6 C. M. Blades. 57-6 3-4 39-3 314 1 Chattaway, Pearmain, \ and Moor. Cream (York), . . . 63-1 1-4 6-5 2-76 Do., do. Dunlop, 3S'5 3-8 31-8 4-1 Johnstone. Dutch, mean of two") samples, . . . . / 39G 6-4 11-5 4-8 ■ Chattaway, Pearmain, and Moor. Gloucester (Double), \ mean of two samples, J 35-2 4-8 25-8 4-7 Do., do. Gloucester, .... 35-8 4-2 21-9 6-2 Johnstone. Gloucester (Single), 21-4 41 25-4 7-7 A. Wynter Blyth. Gruyere, mean of five") samples, . . . . / .34-7 3-S 28-9 4-9 / Payeu, Lindt, and C. t Muller. Gruyfere, mean of twol samples, .... J L"69 4-2 30-2 4-7 ( Chattaway, Pearmain, \ and Moor. Fromage de Brie, . . 51-9 5-0 24-S 2-9 A. Wynter Blyth. Neufchatel, .... 37-9 3-4 41-3 3-7 Do. Parmesan, ..... 32-5 6-2 17-1 6-9 J Chattaway, Pearmain, 1 and Moor. Roquefort, .... 296 6-7 30-3 4-45 Do., do. Skim cheese, .... 431 6-2 0-9 7-7 A Wynter Blyth. "Wen sley dale, . . . 28-3 3-7 38-3 4-3 ' Chattaway, Pearmain, and Moor. ^§ 190, 191.] cuEESE. 381 § 190. Parmesan Cheese is a peculiar cheese, never made iu tbis country, but imported from Parma and elsewhere. The essential points in the manufacture are, that the rennet is heated to about 46°-G (11G° Fahr.), and an hour afterwards the milk set over a slow fire until heated to about 65°-5 (150° Fahr.) These operations cause the curd to separate in hard lumps. It is usually coloured with saffron. The outer crust of the cheese at the end of fourteen days is cut ofiP, the new surface varnished ■with linseed oil, and one side coloured red. It is a very dry cheese, with a large amount of casein, and ouly a moderate percentage of fat. § 191. The Rijyening of Cheese. — The transformation that cheese undergoes, and by which it usually acquires a more agreeable taste and flavour, is without doubt a fermentation of a slow character, induced by the agency of minute mycoderms; possibly, as F. Cohn suggests, the very active thread-bacteria, which rennet always contains, have something to do with the process. Fresh cheese has an acid reaction, but the development of ammonia ultimately renders the reaction alkaline. There is ■with age a continuous loss of water, and a slow development of carbon dioxide— the fat decomposes, and the fatty acids* unite in part with the lime and in part -with the ammonia. The casein also gradually passes into a soluble condition, so that a cheese, which originally gave very little to ■water, ultimately becomes almost -wholly soluble. Blondeau and others considered that, in the ripening of cheese, the albuminoids slowly changed into fat ; thus a Roquefort, a cheese made from ewe's milk, analysed by Blondeau,! had when fresh — Per cent. Casein, 85-43 Fat, l-'85 Lactic acid, 88 Water, 11-84 The same cheese, in the condition most highly prized (after it had been kept two months in a cold cellar), had the following composition : — Per cent. Casein, 43-28 Fat ^Margarin, 18-30 ) 09. on ^^*' lOlein, 14-00 r—^" Butyric acid, 0-67 Chloride of sodium, ...... 4-45 Water, 19-30 * Valerianic acid has been detected in a Roquefort cheese by M. Balard, and by Messrs. J. Genko and Laskowski in a Limbourg cheese, t Annates de Chimie et de Physique, (4), t. i., 1864. 382 FOODS : their composition and analysis. [§ 191. The cheese was again analysed at the end of a year, and its composition found to be as follows : — Per cent. Casern, 40-28 Margarin, . . . . . . . . 16 '85 Oleiu, 1-48 Butjrate of ammonia, , . , . . 5 "62 Capioate of ammonia, 7"31 Caprylate of ammonia, ..... 4"18 Cap rate of ammonia, . . . . . . 4 '21 ChJoride of sodium, 4-25 Water, 15-62 Alexander Midler has, however, shown that the albuminoid and fatty constituents of old cheese bear the same relation to each other as in new cheese, which he proves by calculating the determinations according to an equal quantity of water, thus — Fresh. A year old. Water, 40-4 40-4 Casein, 2i-5 24-5 Fat, 28-0 28-2 Sugar, 1-7 2-6 Salts, 5-4 4-3 N. Siber* has also contested the change of albuminoids into fatj and gives the following analyses of a Eoquefort : — After Very Fresh Cheese, one month. old cheese. Water 49-66 36-93 23-54 Casein,. . . . 13-72 5-02 8-53 Soluble albuminoids, . 6-93 20-77 18-47 Fat, .... 27-41 31-23 40-13 Ash, . . . . 1-74 4-78 6-27 99-46 98-73 9694 Now, although there is in the above analyses an apparent increase in the fat, yet, if reckoned on 100 parts of the dry- substance, there is no very decided change. Thus — Fat. Albuminoids. Fresh cheese, 53-91 40-80 After one month, .... 49-94 40-53 Very old cheese, .... 56-14 37-78 Brassiei",t some years ago, made several careful analyses, which may be of assistance in following the changes that cheese under- goes througli age. Five pieces of the same cheese in the salted and nnsalted condition were analysed at the end of two, four, and seven months, the results of which are tabulated in Table .XXI. The development of ammonia, the ])roduction of nitro- * Jourii. f. Pralc. Chemie, xxi. 213. t Ann. de Chem. Phijs., v. §192.] 383 genous pi'oducts soluble in alcoliol, and the wasting of the fab and total nitrogen, are well shown in these analyses. TABLE XXI. Unsalted pieces. Salted pieces. Fresh cneese. After After After After After 2 months. 4 months. 2 months. i mouths. 7 months. GlTIl. Grm. Grm. Grm. Gi-m. Grm. Casein, . . . 96-21 83 10 85-01 78-60 SU-10 07 06 Milk-sugar, . . 11-46 ... Leucin and other nitrogen- ous substances } ••• 21-18 18-67 15-75 18-28 33-42 soluble in al- cohol, . . . Fat, .... 66-78 56-31 46-92 56-01 40-50 39-74 Insoluble salts, . 2-25 2 25 2-25 Ammonia, . . l'5-53 16 -75 .l'6-50 Soluble sub- stances, . . Trace 1-85 1-95 1-42 1-67 3-22 Water, . . . 123-3 67-31 59-20 68-09 81-70 56-06 Total weight, . 300 232-0 214-0 236-0 239-0 216-0 Total nitrogen, . 15-27 15-94 12-32 13-73 14-63 10-58. The researches on the " ripening" of cheese hitherto made hy no means exhaust the subject, and there appears room for much interesting work in tliis direction. § 192. The Analysis of Cheese. — The chief difficulty in the analysis of cheese is in the extraction of the fat ; one method, recommended and used by the American chemists, is to take a weighed quantity of cheese, mix it intimately with anhydrous copper sulphate, and then exhaust it in a Soxhlet with petroleum, ether, or other suitable fat solvent. A second method is to rub up 25 to 50 grms. of the cheese with sand or gypsum, and similarly exhaust with ether or petroleum. The fat in cheese may also be estimated by " whirling." Chattaway, Pearmain, and Moor operate as follows, using a LefFmann and Beam apparatus : — 2 grms. of cheese are reduced to as fine a division as possible, transferred to a small dish, and heated in the water- bath with 30 cc. of concentrated hydrochloric acid until dis- solved. The mixture is poured into a LefFmann-Beam bottle, the dish rinsed with the hydrochloric acid and fusel-oil mixture into tlie bottle, and, finally, enough strong hot acid to fill the S84 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 192. bottle lip to the mark. It is then " whirled." The Leffmann- Beam bottles are graduated, so that ten divisions equal 1 per cent, by weight of fat in the 15-55 grms. of milk taken ; hence, the factor to be used is -^ — = 7-7. The nitrogen is usually estimated by the Kjeldahl process. The water in cheese is best estimated by extracting the powdered cheese by alcohol and ether, and drying the alcohol €ther extract and the fat-free solids separately. This is a better method than drying the solids in the ordinary way, which will be found in many cases to be extremely tedious and inaccurate. With alkaline cheeses there appears no other way of obtaining the amount of water than careful neutralisation of the free alkali by an acid before drying. The casein and albumen represent the insoluble portion of the cheese. The sugar must be determined on the ])rinciples laid down at p. 287; the lactic acid and the ash on the principles laid down at pp. 286, 290, et seq. Adulterations. — The chief adulteration is the incorporation of animal ftit — this is efiected as follows :* — "An emulsion of lard is made by bringing together in a disintegrator lard and skimmed milk, both previously heated to 140° Fahr. in steam jacketed tanks ; the disintegrator consists of a cylinder revolv- ing within a cylindrical shell ; the surface of the cylinder is covered with fine serrated projections, each one of which is a tooth with a sharp point ; as this cylinder revolves rapidly within its shell the mixture of melted lard and hot skimmed milk is forced up in the narrow interspace and the lard becomes very finely divided and most intimately mixed or emulsionised with the milk. This emulsion consists of from two to three parts of milk to one of lard ; it can be made at one factory and taken to another to be used for cheese, but it is usually run at ■once into the cheese vat. " In making the cheese a quantity of this emulsion, containing about 80 lbs. of lard, is added to 6,000 lbs. of skimmed milk and about 600 lbs. of butter milk in the cheese vat, and the lard that does not remain incorporated with the milk or curd (usually about 10 lbs.) is carefully skimmed off. These quantities of materials yield from 500 to 600 lbs. of cheese, containing about 70 lbs. of lard, or about 14 per cent. About half of the fat removed in the skimming of milk is replaced by lard." Hence, the fat extracted from cheese should always be tested by the Valenta test and by the Reichert-Meissl process, and if a suspicious result is obtained, it must then be analysed in the *U. S. Dep. of Agriculture, 5uW., No. 13, 1887, "Foods and Food Adulteration," H. \V. Wiley. § 192.] CHEESE. 385 same way as butter. In such a case it is best to extract the fat from 100 to 150 grms. of the cheese, so as to have sufficient for the various methods described under " Butter." Other adulterations of cheese which have actually been found are not numerous. All mineral adulterations, save those of volatile metals, must be looked for in the ash, which consists normally of common salt, alkaline, and earthy phosphates. Cheese has from the earliest to the present tim'e been coloured by vegetable matters, and so long as the latter are not injurious, such addition cannot be considered as adulteration. Thin slices of cheese should be examined microscopically after dissolving out the fat, &c., by ether ; in this way starches and vegetable substances may be detected. Arsenical washes and lead pastes have often been applied to the rind to prevent the attacks of the fly. As this part is habitually eaten by a few people, it is necessary to examine it, especially for these metals, and, in a complete investigatk)n, to make two separate analyses, one of the rind, and the other of the substance of the cheese. In past times, a few isolated instances have occurred in which it was found that the manufacturers of cheese had mixed pre- parations of arsenic with the cheese itself as a preservative : — e.g., such was the case in the year ISll, when several of the inhabitants of Chatillon were poisoned by this means. In 1854, the same thing occurred, and a Parisian family sutfered, but not fatally {Chevallier). It is to be hoped that such ignorance is a thing entirely of the past.* Tyrotoxicon, fully described at p. 307, has been discovered by Vaughan ("Report of Michigan State Board of Health," 1886) in a cheese which caused the illness of some 300 people in Michigan. The crystalline substance (diazobenzene) was isolated from a watery extract of the cheese by the process detailed (p. 307). * Whey. — After the removal of the curd from milk, a "whey" remains, containing albumen, peptones, milk-sugar, lactic acid, and salts. The mean of thirty-two analyses collected by Kbnig gives the following as the average composition of "whey": — . Water, 93-31 Nitrogenous substances, -82 Fat, -24 Milk-sugar, 4-65 Lactic acid, ....... -33 Salts, -65 Whey, on account of its high content of milk-sugar, is used for fhe pre- paration of the latter, and there is also a "whey -vinegar." In very many places, however, whey is used merely as a food for pigs. 26 386' foods: their composition and analysis. [§ 192a. BIBLIOGRAPHY. Blades, Charles M.— Cheshire Cheese. Analyst, June, 1894. 'Bi.O'S'DEAV.-'Annales de Vhim. ei de Physiq%ie, (4), i., 1864. Blyth, a. Wynter. — Art. Cheese, "Dicty. of Hygiene." Brassier. — Ann. de Chim. Pliys., v. 270; Chem. Centr. BL, 1865. Chattaway, W., T. H. Peaemain, and J. C. Moor. Analyst, July, 1894, June, 1895. CoHN, F. — D'mgl. Polyt. J., ccxx. 191 ; Journal ofCliem, Society, ii., 1877, 342. Kellner, 0. — Landwirthsch. Versuch, 26, s. 39. MtJLLER, A. — Landw. Jahrbiicher, 1872. Muspratt's " Chemical Dictionary," Art. Cheese. Storch, Y.—Bied. Cenirbl., 1880,366-370. SiEBER, N. — " Ueber die angebliche Umwandlung des Eiweiss im Fett beim Eeifen des P^oquefort Kase." Joicrn.f. prah. Chem., xxi. 213. Vaughan. — "Report of Michigan State Board of Health," 1886. ViETH, P. — Composition of Cream Cheese. Analyst, 1886, 162. LARD. § 192a. Lard is, from an analyst's point of view, the fat of the pig. Lard used to be defined as the fat from particular parts of the pig — that is, the solid fat from around the kidneys, and from the peritoneum ; but the enormous American industry in lard has thrown on the market pig-fat from all portions of the animal, and the term "lard" cannot properly be restricted to fat from particular parts. On the other hand, "leaf-lard " should always be fat from what the butchers call the "leaf" — that is, the peritoneum. The American packing trade, according to Wiley,* " render " the following varieties : — 1. Neutral Lard. — Fat derived from the leaf of the pig rendered in a fresh state at a temperature of from 40° to 50°; this lard is not much exported, but is mainly used up in the manufacture of oleo-margarine. It is almost neutral. 2. Choice Lard, Choice Kettle-rendered Lard. — This is lard made from leaf and trimmings only. It is rendered in steam- jacketed open kettles, hence its name. 3. Prime Steam Lard. — This is apparently pig- fat rendered by steam from all or any portions of the animal. 4. A low quality lard made from the "guts," by which term the hog packers mean the whole of the abdominal viscera. It is obvious, therefore, that different lards will show analytical variations. Spaeth has made some useful observations on the chief analytical and physical distinctions of lard derived from various portions of eight animals. His results are as follows : — * U. S. Dep. of Agriculture, " Foods and Food Adulteration," by H. W. Wiley. 192i.l 587 Fat from the Back. Fat from the Kidney. Fat from the Leaf. Specific gravity at 100° (water at 15" =1), Melting-point of fat, . ,, of fatty acids, 0-8607 33° -8 40° -0 0-8590 43° -2 43° -2 0-S58S 44° -5 4-2° -9 Iodine value of fat, . ,, of fatty acids, 60-58 61-90 52-60 54-20 53-10 54-40 Free fatty acids, KOH per 100 grms., . ,, ,, calc. as oleic acid, 0-54 0-152 0-58 0-163 1-28 0-360 Lard consists of the glycerides of palmitic, stearic, and oleic acids. Linoleic acid has been stated to be a constituent of lard,, but this requires confirmation. § 19'2b. The Physical Characteristics of Lard. — Lard possesses a- pure white colour and a granular texture. The specific gravity of lard at 40° -0 is about 6-898 ; at lUO" about 0-861— that is, ia each case, water at IS^-O being taken as 1. The increase in density for each degree C. is -00062. The melting-point of lai'd is from 34° -0 to 48°; Wiley, for American lards, considers a wide deviation from 40° as a sign of impurity. The solidifying j^oint of butcher's lard'is 28°*6 to 29°-9. American steam lard from 25° to 27°. A better test is the solidifying point of the fatty acids ; this, in the prime steam lards, varies, according to Wiley, from 41°-4 to 43°-0; in other lards from 36°-9 to 46°-6; these numbers agree fairly well with those found by Gladding. The refi'action of lard has been studied by Wiley, using an Abbe's refractometer. The rate of variation for each degree of temperature was found to be 0-000288, and the refractive index varies inversely as tlie temperature. The refractive indices were taken at various temperatures between 30° and 40°, and then reduced to 25° ; at that temperature water had a refractive index of 1-3300, and the mean number for pure lard was 1-4620. Microscopical Appearances of Lard. — If forty drops of melted lard are dissolved in 10 cc. of ether and allowed to cool, crystals are soon deposited, which have the appearance of oblong plates with oblique terminals ; these occasionally occur in radiated groups. Beef-fat, on the other hand, always forms more or less radiated groups of crystals, and the individual cx-ystals are in the form of 388 foods: their composition and analysis. [§ 192c. needles, not plates (see Frontispiece). Some of the crystals are of a wavy /shape. § 192c. Chemical Characteristics of Lard — 2Ioisture in Lard. — The quantity of water in lard is fractional ; it should never exceed 0-7 per cent. An amount of water equal to 1 per cent, would be an adulteration. Lnsoluble Fatty Acids. — These vary from 93 to 95 per cent. Volatile Acids. — The volatile acids are always small in quantity. If a Reichert test be applied and 5 grms. be saponified, decom- posed, and distilled, the distillate does not neutralise more than 0-5 cc. of d. n. soda. Should the amount of volatile acid be more than this, it is a suspicious sign. Saponification Value. — This has been stated by Koettstorfer to be equal to 195-6 mgrms. of KHO, by Valenta as from 195-3 to 19G-ti, and by Wiley as varying from 193-4 up to 203-1 with a mean value of 198. Iodine Number. — This is a value depending on the amount of olein in the lard ; it I'ises above the normal when lard is adulterated with oils of larger olein content and sinks below the normal when lard is mixed with stearins, having a small iodine absorption. Mixtures can be made, however, which will pass the iodine test. Unfortunately the normal iodine number varies much in the fat of the pig derived from different parts of the animal ; compare for instance the iodine values given at p. 387. Wiley, for prime steam American lards, gives the value as from 60-34 to 66-47 per cent, of iodine absorbed, with a mean value of 62-5 ; Hubl gives a mean number of 59 ; Dieterich from 49-9 to 63-8. So that all that the analyst can say is that a number below 49 or higher than 64 is a suspicious sign. Another method is to separate the insoluble fatty acids, and as in Muter and de Koningh's process (already described, p. 369) determine the iodine number of the liquid fatty acids. Pure lard, according to Muter, yields a liquid fatty acid having an iodine number of 94 ; cotton-seed oil, on the other hand, has an absorption under the same conditions of 136 ; so that if the iodine number is above 94 adulteration with vegetable oil is indicated. Muter's iodine number of 94 must not be considered final, for Wallenstein and Finck* have made some researches on the liquid fatty acids of European lards, and give the iodine number as from 93 to 96, and the iodine number of the liquid fatty acids of American lards as high as from 103 to 106. 2'Jie Maumene Test. — This is the rise of temperature when a definite quantity of an oil is mixed with a definite quantity of *Chem. Zeit., 1894, 1189. § 192c.] LAKD, 389 strong sulphuric acid. Tiie drying oils examined in this way give a much higher temperature than the non-di'ying oils. Comparable results are obtained by operating on strictly the same conditions. These conditions are to use the same volumes or weights of oil and acid, to bring the two to the same initial temperature before mixing, and to have the same strength of acid. This latter may, however, not be essential if Thomson and Ballantyne's suggestion be adopted of referring the rise of temperature obtained with 50 grms. of oil and 10 cc. of sulphuric acid, to the rise of tempera- ture which 50 grms. of water give under the same conditions ; the rise of temperature with the oil, divided by the rise of temperature with the water, gives a quotient which they have named " specific temperature reaction." This in some degree meets the difficulty of testing with different strengths of acids. The best method of bringing about a temperature reaction after Maumene's method is as follows : — .Strong sulphuric acid is poured into di-y stoppered bottles of small capacity, and all save one securely protected by a lute from the air. The analyst can now get his standard from the small bottle in use, and this standard will hold good for all the rest. To apply the principle to lard, 50 grms. of the lard are melted in a small flask and kept at a temperature some 2° or 3° above the melting-point, by standing in a water-bath kept at a constant temperature. The small bottle of sulphuric acid is also placed in the same water-bath, and when both acid and melted lard are equal in temperature, 10 cc. of sulphuric acid are transferred to the oil at the same moment that the flask is immersed in a beaker and packed with cotton wool; the mixture is then well stirred with the thei'mometer and the rise of temperature noted. Such various degrees of heat have been stated that it would be unwise to rely upon any ])ublished statement, but a standard should be made by operating on pure lard ; under these circum- stances good comparative results will be obtained. For example, Engler and Rupp found a lard giving a rise of temperature of 31° to 32°; the same lard with 10 per cent, of cotton-seed oil, gave a reaction of 3-1° ; with 20 per cent, cotton seed oil, 40° to 42°; and with 50 per cent, seed oil, 58°. In all cases, indeed, the same operator will find a difference between the temperature reactions of pure lard and lard mixed with oils of high tempera- ture reactions. Hehner and Mitchell* have proposed to treat the fat with bromine and ascertain the rise of temperature. 1 grm. of the oil or fatty acid is dissolved in 10 cc. of chloroform in a test * AiuihjHt, July, 1895. 390 foods: their composition and analysis. [§ 192(/. tube enclosed in a vacuum jacket ; 1 cc. of bromine is added, and the rise in temperature noted by a delicate thermometer. The bromine, oil, and solvent must all be at the same temperature before mixing. The rise of temperature of 10 samples of lard gave a mean number of 10°-7; a sample of lard with 10 per cent, cotton oil, ir-6; cotton oil, 19°-4; cod-liver oil, 28°; olive oil, 15°; butter, 7°; by multiplying the number of degrees by the factor 5-5 a close approximation to the iodine number is obtained. Thus the process checks a Hubl determination. § 192d Adulteration of Lard. — The chief sophistications of lard are the admixture of beef stearin, either alone or with vegetable oils, especially cotton-seed oil, cocoa oil, sesame oil, or maize oil. Cotton-seed stearin has also been used, either with beef stearin or alone. Water is generally enumerated as an adulterant, but the wilful addition of water at the present time is rare. The adulteration of lard is on a lai'ge scale, and great ingenuity has been shown in devising mixtures which will pass the analy- tical tests. Hence the analyst, when he examines a suspicious sample, must never condemn on one test, but should ascertain what are the chief lard constants before he pronounces an opinion. The chief analytical values of the different adulterants of lard ■are tabulated in Table XXIa, from which the following facts will be apparent : — (a.) Specific Gravity. — All the adulterants save beef stearin raise the gravity of lard ; beef stearin is, on the other hand, about the specific gravity of lard. (h.) The Solidifying- or Freeziny-Point of the Fat. — Cotton-seed oil, arachis oil, sesame oil, maize oil, and even cocoa-nut oil, liave all lower figures for the freezing-point than lard ; on the other hand, beef stearin is somewhat higher, so that a 10 per cent, admixture of beef stearin will have but little effect on the freezing-point of the mixed fat. Cotton-seed stearin also may have the same freezing-point as lard. (c.) The Freezing- or Solidifying- Point of the Fatty Acids. — The same remarks apply to the solidifying-point of the fatty acids, but here another valuable distinction comes in — viz., the very great difference between the freezing-point of some of the oils and fats and the freezing-point of their fatty acids ; for example, pure lard giving a solidifying-point of 25° to 29°, the fatty acids are on an average 10° to 15° higher; on the other hand, cotton- seed oil as compared with the fatty acids will show 30° of differ- ence, while maize oil and cocoa-nut oil will differ but to a slight, often inappreciable, extent. § 192cZ.] LARD. 391 -.1 3 a « ? ^1 in 00 .S 1^ 1 k S 1 IS - . i i s i lb 11 o in 3 ! ! ! s ! ! O CO O — 1 CS CO Ills 2 1 2 2 § 11 -H O CO ■03 O CO • 03 C5 CO 00 Tj< 05 o p o o o pop «0 O O O CO CO «o 05 C5 Oi O 00 CS 05 1 1 CO k 1 li s § 9 p j^ °^ !2 °2: 2 - - ^ ^ ^ o o o IQ O t^ CI — 1 t- O ^ CM CO CO o lO th o o in to 1; •Sfe; 1 i S § rt § s <» 00 CO 00 CO o ?> t^ '*< CO OO <» CO lO 00 00 CO 00 <» 1 s i ^ g p p p p ^ p „ gi 2 ?t ?3 "?^ ^ ^ 9 j 1 § i 3 2 ■g i i. 3 2 2 1 g "g "S % •a >> >. < " < < " ■ ■< Chloroform, . . 12-97 1/7-72 19-02 1/5-25 Alcohol of 85 per \ cent., ... J Water, .... 2-51 2-30 1-40 1/44-4 1-47 1-.35 1-68 1/74-2 49-73 45-55 1/2-01 1/2-19 Absolute Alcohol, 0-61 1/164-7 312 1/32 Commercial Ether, 0'-'21 0-19 1/476 1/526 ... 0-454 1/220 Carbon Bisulphide, ... 0-0585 1/1709 ... ... Purified Anhy- "1 drous Ether, J ... 0437 ... 1/2288 ... 0-36 1/277 Light Petroleum, . 0-025 1/4000 the silver compound, separated from a concentrated -watery- solution, and the mercury compound, almost insoluble in ether, and capable of being dried at 100°, might possibly be of use in the estimation of theine. Theine is, in large doses, a poison. Frerichs, C. J. Lehmann, Husemann, and others, have made themselves the subject of experiment. Lehmann, after taking -5 grm., suffered from fre- quency of the pulse, irritation of the bladder, cerebral excitement, slight hallucinations, and, lastly, desire to sleep. Husemann took -25 grm. with somewhat similar symptoms. Pratt, with subcutaneous injections of from -12 to -8 grm., suffered from symptoms rather different from the foregoing ; -3 grm. lessened the pulse and caused sleeplessness ; -4 to -5 grm. quickened the- pulse, and caused a desire for frequent micturition, but no- dilation of the pupil ; -8 grm. caused gi'eat uneasiness and anxiety, trembling of the hands and arms, so that he was. unable to write, and later a restless sleep, with continual dream- ing. In opposition again to all these statements, is that of the late Mr. Cooley,* who is said to have taken 20 grains (1-29 grm.) of pure theine every day for a month without experiencing any other symptom than some slight elevation of spirits. According to Strauch, the least fatal dose for cats is -25 grm., a quantity * Vide Cooley's " Dictionary of Practical Recipes," Art. Caffeine. 400 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 195, which killed a cat in thirty-five minutes. In all experiments on animals there has been increased frequency of the heart's action, and repeated emptying of the bladder and intestine. No case of poisoning in the human subject appears to be on record. When given to animals it has been chemically separated from the blood, urine, and bile. Tests for Theine. — Concentrated sulphuric and nitric acids •dissolve theine in the cold without the production of colour. If the alkaloid is treated with fuming nitric acid, and evaporated to dryness, the reddish-yellow residue becomes, when moistened with ammonia, of a splendid purple-red colour. If a solution of theine be evaporated with chlorine water in a watch-glass, a red- brown residue is obtained, which on cooling, and exposure to the vapour of strong ammonia, becomes purple-violet. The chief precipitants of theine are — phospho-molybdic acid, yellow preci- pitate ; iodine with potassic iodide, dirty brown precipitate ; chloride of platinum, yellow hair-like crystals, insoluble in cold hydrochloric acid, slowly separating ; chlorides of gold, mercury, and nitrate of silver also give precipitates. BoIipAg Acid, CyH^oOg, was first separated by Eochleder in 1847,* fi'om the leaves of Thea sinensis. The hot watery decoc- tion of tea is precipitated whilst boiling by sugar of lead, filtered, the filti'ate neutralised by ammonia, the resulting precipitate collected, suspended in absolute alcohol, and freed from lead by SH2 ; the filtrate from the lead precipitate is evaporated to dry- ness in a vacuum, and purified by re-solution in water', &c. It is a pale yellow amorphous powder, melting at 100° into a tenacious mass, and decomposing at common temperatures if exposed to the air. It is soluble in all pro})ortions in water and alcohol, is coloured brown (but not precipitated) by chloride of iron, and forms for the most part amorphous salts insoluble in water. Quercitannic Acid, CogHgoOj^, first discovered by Chevreul and Brandt in the Quercus tinctoria, and stated by Hlasiwetz to be in tea leaves, can be crystallised from an aqueous solution. It forms sulphur or chrome-yellow microscopic tables, containing -3 atoms of water, part of which is expelled at 100°, the rest at from 165° to 200°. Its reaction is neutral, and it is without odour, but has a marked bitter taste when in solution. It melts at from 160° to 200° to a resinous, amorphous mass. Its solu- bility is as follows : — Cold water 2485, boiling 143 ; cold absolute alcohol 23 -3, boiling 3-9 ; ether dissolves it slightly, warm acetic acid copiously. Sugar of lead precipitates almost completely; the precipitate is soluble in acetic acid, * Rochleder: Ann. Cheni. Pharm., Ixiii. 202. COMPOSITION OF TEAS IN BUSSIAN COlvIMEBCE. M Water. Par cent. ThBine in ss- ■—?■ Per cent 53-6 51-8 PetSnt Percent °Sr Art from insoluble Per cent "' .S.. Potash, [O^fheAji,.] Pbosphorio PhOBphorio Percent [Od the Ash.] 1 1. YeUowTea, . 2. 10-90 7 10 35-5 411 1-78 1-61 672 5-92 6-12 1-64 0-35 5-28 5-77 2-16 1-99 2-53 42-82 109 18-47 j LeDgth of leaf; 30 to 40 mm,; only a few entire a „ „ 4. 909 9-88 38-8 36-6 1-49 1-43 6-60 12-70 52-2 53-7 5-61 6-33 0-27 0-85 5-34 4-48 2-12 2-79 2-10 39-55 0-80 1-33 16-65 25 02 (Many broken stalks, with folded undeveloped eaves; buds 3 to 5 mm, diameter; length of I Buds knd points of leaves still less developed; but 1 few entire leaves. 5. Green Tea, . 6. ., .. 7 2-50 4-50 8-35 8-82 33-5 39-9 37-5 1-82 1-66 1-61 602 6-01 12 32 58-2 51-3 54-6 6-82 6-21 5-78 0-85 0-98 0-58 5-97 5-23 5-20 2-74 2-50 2-04 2-27 2-48 .-J3-40 42-99 0-79 0-72 0-77 U-65 13.32 \ Full grown, but not leathery leaves; most, it would I appear, halved by desiga; absence of buds. No. > 5 more compact than the others ; coloared blue, and slightly rolled. No 6. is in roUed little J balls; No, 7 in cylinders. 8. Black Tea, . 1-20 10-63 44-5 1-36 4-15 449 6-07 0-99 6-08 2-89 2-23 .•i6-88 0-79 13-11 Mostly old leaves; much divided. 9. ., „ 1-40 10-25 32-4 1-79 57-4 6-51 0-83 5-68 3-68 0-93 14-30 10. 1-60 1-75 10-43 9-98 33-3 26-8 1-65 1-89 ... ... 56-3 63-3 6-00 6-22 1-35 0-89 4-65 6-24 3-75 3 04 0-83 13-84 OM leaves, but little powdered ; mostly in halves. \ 11 and 12 are less compact than the rest, so f that by a little soaking in water they are easily powdered. 12. „ 9-47 30-7 2-08 ... 9-42 59-9 5-62 1-19 4-43 3-19 1-03 18-42 13. „ „ 14. „ 2-20 2-50 10-70 10-90 27-2 27-2 2-11 2-14 5-26 ... 62-1 61-9 6-18 5-78 1-11 0-98 5-70 4-80 2-79 1-97 34-19 1-14 1-11 18-54 17-25 ' Mostly not fully grown; halved; rather thin leaves. 15. Flower Tea 1 (Blumen Thee) ( 2 '20 9-46 29-1 2-12 67-5 6-16 1-03 5-12 2-69 2-28 .37-94 0-88 14-42 ) Similar to the above; a few stalks as Nos. 3 and 4; } No, 10 has somewhat more numerous stalks 16. „ 2-50 8-79 300 2-13 4-75 61-3 5-89 1-12 4-77 3-16 0-87 14-79 \ than 15, 17. 2-70 10-51 29-4 1-81 60-1 5-62 0-92 4-70 2-32 1-27 19-92 18. „ „ 19 3-20 12-66 12-00 24-8 26-7 1-79 1-95 62-6 61-3 5-66 6 20 0-83 0-97 4-83 6-23 2-47 2-47 ... •0-94 104 16-71 16-88 >Cut, half-developed leaves; a few stalks and buds. 20. „ 3-50 1109 30-5 1-79 58-5 6-57 0-97 6-60 2-45 0-94 14-43 21. .. " „ . 10-36 31-2 2 02 58-5 5-45 0-89 4-66 2-38 ... ! 1-24 22-83 ) 22. „ „ 10-72 31-9 2-68 11-24 57-4 5 -48 0-73 4-75 204 1-27 23-19 \ Young leaves without buds. 23. 11-05 32-8 3-09 5-19 57-2 5 83 0-54 5-29 1-99 2-23 .38-24 1-56 25-64 j § 190.] COMPOSITION OF TEA. 401 Quercetin, first obtained by Rigaud, 1854, from the splitting up of quercitannic acid is, according to Filhol, to be found in the green leaves and flowers of all plants. Its formula is given as Co^H-^gOjg ; it forms fine yellow needles, or a citron-yellow powder, which gives up at a temperature of 120°, 7 to 10 per cent, of water of crystallisation. It melts, according to Zwenger and Dronke, above 250° without decomposition, solidifying again in a crystalline mass, and it may be also sublimed with only partial carbonisation. It is very little soluble in water. "Warm acetic acid dissolves it copiously, but it separates on cooling. It is soluble in 229-2 parts of cold, and 18-2 parts of hot absolute alcohol. A solution of quercetin colours linen bi'ight yellow ; sugar of lead precipitates the alcoholic solution cherry red, and chloride of iron dark red. A combination with sodium or potassium can be obtained, ISra20,C.,jH^gO;^2- The other constituents of tea, such as gallic and tannic acids, gum, &c., are too well known to need description. COMPOSITION OF TEA. § 196. The mean of sixteen analyses quoted by Konig gives the following : — Water, . Nitrogenous substances Theine, . Ethereal oil, . Fat, chlorophyll, ^ Gum and dextrin. Tannin, . Other nitrogen free mati Woody fibre, Ash, . 100 00 And this may be taken to represent average tea. The annexed Table gives determinations of several constituents of teas in Russian commerce by Dragendorff" — the chief features of which may be siimmaiised as follows : — 27 Per cent. 11-49 21-22 1-35 •67 3-62 7-13 12-36 16-75 20-30 5-11 402 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 196. Bragendorff' s Analyses of Ticenty-three Teas in Russian Commerce. Mean. Maximum. Minimum. Per cent. Per cent. Per cent. Water, 10 00 12-66 7-10 Extract, .... 32-67 44-50 24-80 Theine, .... 1-90 3-09 1-36 Tannin (4 determinations only), 11-42 9-42 12-70 Ash, 6-23 6-82 5-23* An interesting reseai'ch on the changes taking ph\ce in the tea-plant through age has been made by O. Kellner.f Tea leaves from the same plants were collected twice a month from May to November, and a sample also obtained at the end of the twelve months. The main results may be briefly summarised thus : — The water regularly diminished from 76 to 60 per cent. ; the theine in the young leaves amounting to 2-85 per cent, of the dried substance diminished to -84 per cent, of the old dried leaves, and the total nitrogen from 4-91 to 2-67 per cent. The nitrogen from the amido-acids, equal in the first month to '66 per cent, of the dried substance, rapidly decreased, so that in September the amido-nitrogen only attained -08 per cent., and in the twelfth month was still less — viz., -01 per cent.; since the theine did not diminish in anything like the same proportion, the inference is that the amido-nitrogen became theine- nitrogen. On the other hand, the tannin increased from 8 to 11 or 12 per cent., and the ash from 4-69 to 5*14 per cent. It is, therefore, clear that in young tea leaves there are more water, more theine, more amido-acids than in old leaves. Whereas in old leaves, with the decrease of the constituents mentioned, there are more salts and more tannin. * The analyses of tea by Mulder, which appeared in the earlier edition of this work, are as follows : — Essential oil Chlorophyll, Wax, Resin, Gum, Tannin, . Theine, . Extractive matter. Colouring substances, Albumen, Fibre, . Ash (mineral substances). The theine is certainly too low. t Laud. Versuchs. Stat., 1886, 370-380. Journ. Chem. Soc, Jan. 1887, 73. JIack Tea. Green Tea. 0-60 0-79 1-84 2 22 00 0-28 3-64 2-22 7-28 8-56 12-88 17-80 0-46 0-43 21-36 22-80 19-19 23-60 2-80 3-00 28-33 17-80 5-24 5-56 § 197.] THE DETECTION OF ADULTERATIONS IN TEA. 403 MICROSCOPICAL METHODS OF DETECTING ADULTERATIONS IN TEA. PreUminarTj Examination of Tea. — The tea leaves should be soaked in hot- water, carefully unrolled, and their shape and structure examined. Sections of leaves can be made by placing them between two pieces of cork, and cutting fine slices otF both, the cork and the enclosed leaf; on floating the sections in water, the film of cork may be readily separated from the leaf The epidermis of the lower or upper surface can, with a little practice, be detached in small portions by the aid of a sharp razor, and examined in water, glycerin, or dammar balsam under the microscope. Its structure has been already detailed. § 197. New Process for the Examination of Leaves and Vegetable Tissues generally under the Microscope. — The author has dis- covered a very easy process for examining vegetable leaves. A portion of a leaf is enclosed between two of the thin circles of glass used by all microscopists, and a weight having been placed upon the upper glass, the portion of leaf thus enclosed is heated with a strongly alkaline solution of permanganate of potash. The action begins at once, and the substance under examination must be examined from time to time to see that the oxidisation does not proceed too far. Alkaline permanganate attacks the colouring- matters, the contents of the cells first, and afterwards the cell membranes ; the object of this treatment is to make the leaf transparent, and yet to preserve its structure. Tea leaves are very opaque, and it is impossible without some mechanical or chemical treatment to render them transpai-ent.* When froni the appearance of the leaf -fragment the oxidation is considered sufiicient, it is removed, washed in water, and treated with a little strong hydrochloric acid, which at once dissolves the manganese oxide that has been precipitated on the leaf, and leaves the latter as a translucent white membrane, in which the details of structure can be readily made out — tea-leaf in this way being quite different in appearance from other leaves. A second method of very great value is to place a fragment of a leaf between two circles f of glass, weight the upper one with a silver coin, and burn on a bit of sheet platinum the leaf thus prepared. Since it is impossible for the ash to curl up and become disarranged, a complete skeleton of siliceous ash remains, which may be called " the skeleton ash." These skeleton ashes of leaves (so far as the author has hitherto been able to investigate the subject), show such de- * See also method to discover " idioblasts " described at page 397. + That is, the covers of thin glass used by the microscopist. 404 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 198. cided differences the one from the other, that a great number of leaves may with a little practice be recognised by this method alone. [It is particularly useful in detecting the adul- terations of tobacco, the skeleton-ash of the tobacco-leaf being special and peculiar.] It is sometimes well not to burn to an absolute ash, but to leave little bits of partially consumed carbon, forming objects for the microscope of great beauty. To preserve the " skeleton-ash " the two circles may be cemented together, or the edges may be fused by the flame. Figs. 43, 44, 45, and 46 are examples of skeleton ashes, as drawn on the block to scale. ri- 43. -Aftli of Tea Leaf, X 170. 41.— Ash of Sloe Leaf, X 29. Ficr. 45.— Ash of Lime, X 29. Fig. 46.— Ash of Tobacco Leaf (Cigar), x 29. §198. Chemical Method for the Detection of Foreign Leaves in Tea. — A chemical method for the detection of foreign leaves (adulterants) was first described by the writer in June, 1877.* * Micro-Chemistry as applied to the identificatiou of tea leaves. Analyst, § 198.] THE DETECTION OF ADULTERATIONS IN TEA. 405 It is based upon two facts — firstly, that every part of a theine- producing plant — wood, stem, leaf, flowers, and even hairs — contains the alkaloid ; and, secondly, that this can be readily sublimed. The leaf, or fragment of a leaf, is boiled for a minute in a watch-glass with a very little water, a portion of burnt magnesia equal in bulk is added, and the whole heated to boiling, and rapidly evaporated down to a large-sized drop. This drop is transferred to the " subliming cell," described fully in " Poisons,"* and if no crystalline sublimate be obtained, when heated up to 110° (a temperature far above the subliming point of theine), the fragment cannot be that of a tea-plant. On the other hand, if a sublimate of theine be obtained, it is not con- clusive evidence of the presence of a tea-leaf, since other plants of the camellia tribe contain the alkaloid. Finally, there is a negative test which may occasionally be valuable. All fragments of tea hitherto examined contain man- ganese, and there are a few foreign leaves in which manganese is constantly absent. Hence, if a leaf be burnt to an ash, and a fragment of the ash be taken up on a soda-bead, to which a little potassic nitrate has been added, the absence of the green man- ganate of soda would be sufficient evidence that the leaf had not been derived from the tea-plant, while conversely, as in the case of theine, the presence of manganese is not conclusive of tea. Another portion of the tea leaves should be thoroughly bruised, spread on a glass plate, and carefully searched with a magnet for ferruginous particles — the so-called iron-filings, which are occasionally found, especially in Capers and certain species of Congou. It is almost unnecessary to state that the black, irregular masses found in tea, and attracted by a magnet, are not metallic iron, f Their chemical composition is somewhat variable j they all contain magnetic oxide of iron, and many of them in addition phosphate of iron, titanate of iron, quartz, and mica, with a little sand. They are, without doubt, sometimes an adulteration (the author has himself found over 1 per cent.), and sometimes an impurity, for in a few teas mere traces only of this ferruginous sand may be discovered. Any particles of the kind extracted by the magnet should be collected and treated with hot water, which soon disintegrates them; the adherent tea-dust is separated, and the sand dried and weighed. * Poisons: their Effects and Detection. 3rd. ed., 1S95, p. 258. t Mr. Allen appears to have found metallic iron in tea. The test for metallic iron is, that nitric acid, 1-2 specific gravity, dissolves it with the production of red fumes ; it also precipitates metallic copper, if added to an acidulated solution of cupric sulphate. 406 foods: their composition and analysis. [§198. To detect facing, the tea iu its dried state should be mounted as an opaque object.* If it has the appearance of being heavily faced, soaking in warm water will soon detach the fihn ; and indigo, Prussian blue, or similar substances will sink to the bottom, and may be collected and examined. Indigo may be identified by the microscope. Prussian blue may be tested for by warming the deposit with caustic alkali, filtering, acidifying the filtrate with hydrochloric acid, filtering again if necessary, and testing the filtrate with ferric chloride. The residue left after treatment with caustic alkali may be tested for magnesium silicate, by first extracting with HCl, and then collecting the insoluble residue, and fusing it with an alkaline carbonate. The silica is now separated in the usual way by evaporation with HCl to dryness, subsequent solution in weak acid, and filtration; any lime is removed by ammonia and ammonic oxalate ; and lastly, magnesia is precipitated as ammon. mag. phosphate. Magnesia found under these circumstances must have been present as steatite or other magnesian silicate. * The facing of tea is thus described by M. S. Julien: "The leaves are mixed either with powdered indigo, with powdered plaster, or with slaked lime, sometimes even all thi-ee substances being put together in small pro- portion to tea leaves. These matters are introduced into the basins at the commencement of the operation, when the leaves begin to be covered with a light dew under the influence of heat. These matters attach themselves to the leaves, and communicate to them the bluish-green characteristic of gi-een tea. ... In certain manufactories Prussian blue is used instead of indigo. " " Industries Anciennes et Modei-nes de I'E iipirc Cliiiaois," par MM. Stanislaus Julieu et 0. Champion, Paris, 1869. 199.] ADULTERANTS OF TEA. 407 LEAVES USED, OR SUPPOSED TO BE USED, AS ADULTERANTS. § 199. The following is a brief description of the principal leaves supposed to be used as adulterants : — Beech [Fagus sylvatica). — The leaves of the beech are ovate, glabrous, obscurely dentate, ciliate at the edges, the veins running parallel to one another right to the edge. The leaf, slightly magnified, is seen to be divided into quadrilateral spaces by a network of transparent cells. On section, the parenchyma of the leaf is found to consist of an upper layer of longitudinal cells, and a lower of loose cellular tissue, enclosed between the epidermis of the upper and under surface. The whole section is thus divided into oblong spaces by trans- parent cells connecting the cuticle of the upper and lower sur- faces. The epidermis of both the upper and lower surfaces is composed of cells with an extremely sinuous outline (see fig. 47). The stomata are small, not numerous, and almost round. Beech leaves contain manganese. T^'n^ Fig. 47-— Epidermis of Beech Leaf, x 300. Hawthorn {Cratcegus oxyacantha). — At least two varieties, the more common of which is the G. monogyna, with obovate tliree- to four-deeply lobed leaves, with the lobes acute. The leaf is 408 foods: their composition and analysis. [§199. divided into quadrilateral spaces, like the beech and many other leaves, by a transparent network. The epidermis of the upper surface is composed of a layer of thin-walled cells, generally quadrilateral, outline seldom sinuous. The epidermis of the lower surface has a layer of thin- walled cells, with a very sinuous outline. Stomata large, distinct, and numerous, in many instances nearly round, but the shape mostly oval. (See fig. 48). Fig. 48.— Epidermis from the Under Surface of the Hawthorn Leaf, x 300. Camellia Sassanqua. — The leaves of Camellia sassanqua are oval, obscurely serrate (the younger leaves entire), dark green, glabrous, of somewhat leathery consistence ; the lateral veins of the leaf are inconspicuous. Micro-structure. — The parenchyma of the leaf is placed between two thickened epidermal layers ; the epidermis of the upper sur- face, as seen upon a section, forms a wrinkled, continuous, thick membrane, in which a cellular structure is not very evident. Below this there are two or three layers of large cells, more or less oblong, with their long diameter at right angles to the surface of the leaf; and underneath this again is a loose network of cells, resting upon an epidermis in every respect similar to that of the tipper surface, but only half as thick. A thin layer of either the upper or lower epidermis shows a peculiar dotted or reticulated appearance, not unlike the rugse of a stomach. The lower epidermis is studded with frequent stomata, small, and of an oblong shape (see fig. 49). Sloe {Frunus communis). — The leaves of the common sloe are § 199.] ADULTERANTS OF TEA. 400 Fig. 49.— Epidermis of Under Surface of the Leaf of the Camellia Sassanqua, X 300. rather small, elliptic or ovate-lanceolate in shape, and slightly- downy beneath. The sectional thickness of the leaf is the same Fig. 50.— Section of Sloe Leaf, x SCO. 410 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 199. as that of tea. The stomata on the lower surface are scanty. The microscopical appearances are wholly different from those of tea. leaves, more especially as seen in section. (See fig. 50). Clilorantlius Inconspicuus. — The leaves of the Chloranthus in- consjncutts are long, oval, serrate, wrinkled, the veins running nearly to the edge, and there forming a network in such a manner, that at the point of intersection little knots are foroied, which give the margin of the leaf a very rough feeling. The structure >!.—(«) Epidermis of Under Surface of the Leaf op Chloranthus Inconspicuus, x 300. [b) Section near Edge, § 200.] THE CHEMICAL ANALYSIS OF TEA. 411 of the leaf is veiy simple. The epidermis of the upper sui-face is formed of one or two layers of thin-walled cells, the epidermis of the lower of one or two la3'ers also of cells, and between the two there is a parenchyma of loose cellular tissue. The stomata are oval and rather numerous; their length is from -0010 inch, their breadth -00073 inch. The cells of the epidermis are large, some of them -005 inch or more in their long diameter.* (See fig. 51.) THE CHEMICAL ANALYSIS OF TEA. The sample is next submitted to chemical analysis. If the question to be decided is simply that of adulteration, the taste of the infusion, the percentage of extract, and a determination of the chief constituents of the ash are in most cases all that is necessary ; but a more or less complete examination embraces a quantitative estimation of hygroscopic moisture, theine, total nitrogen, tannin, extract, gum, and ash. § 200 Hygroscopic Moisture. — The ordinary method of taking the hygroscopic moisture of tea is to powder as finely as possible an indeterminate quantity of from 1 to 2 grms., and to heat it in a watch-glass over the water-bath until it ceases to lose weight. It should be finally weighed between two watch-glasses, since it rapidly absorbs moisture from the air. The method given is in its results incorrect, since some volatile oil and a small proportion of theine are always volatilised. That theine is actually lost is capable of rigid demonstration ; it is only necessary to heat a few leaves of tea between two watch- glasses over the water-bath, and theine crystals can be readily discovered on the upper glass. To devise a process of di-ying tea which will represent water only is easy; but since the loss both of volatile principles and theine does not materially affect the results, it is scarcely worth while to complicate the analysis by the use either of a lower temperature or of processes of absorp- tion. The highest amounts of moisture in a genuine tea which are on record are two specimens from Cachar, analysed by Pro- fessor Hodges — the one (indigenous) gave 16*06 per cent., the other, a hybrid, 16*2 per cent. These were, however, not com- mercial teas, and appear to have been simply dried in heated rooms. The average hygroscopic moisture found by Mr. Wigner * The leaves of Epilobium angimtifolium (common lolllow herb) are said to be extensively used in Russia for the adulteration of tea. The dried leaves are sold for from four to six roiables a pouud, and are used bj^ the poorer classes in the place of tea. Alcohol produces in infusions of epilobium a precipitate of mucilage. — Pliarm. Zeitsch. filr Eussland and Ytar-Book of Pharmacy, 1876. 412 foods: their composition and analysis. [§201. in thirty-five teas, consisting of Hysons, Capers, Souchongs, Gun- powders, and others, was 7-67 per cent., the driest teas being the Hysons and Gunpowders, the moistest the Congous : — Per cent. The nitaximum amount of moisture found in Hyson, . 5 "68 The minimum ,, ,, ,, .4-84 The maximum ,, ,, Gunpowder, 6 55 The minimum ,, ,, ,, 4-94 The maximum „ ,, Congou, . 10 "33 The minimum ,, ,, ,, . 6"36 § 201. The Estimation of Theine or Caffeine. — The modern pro- cesses for extracting theine fall chiefly under three heads :— (1.) Extraction hij treating a decoction of the theine-containing substarices with lime or burnt magnesia, evaporation to dryness, and subsequent solution of the alkaloid by chloroform, ether, or benzine. — The fundamental idea of this process, perhaps, belongs to Miiller ; it has also, with various modifications, been recommended by Clous, Commaille, DragendorfF, and many other chemists. Commaille adopts the following method : — 5 grms. of the finely powdered and carefully sifted substance are made into a hard paste with 1 grm. of calcined magnesia. This, after standing for twenty-four hours, is dried upon a water-bath and powdered. The resulting green powder is exhausted three successive times in a flask with boiling chloroform, the flask being connected with an inverted Liebig's condenser, so that the action may be continued for a long time. The cool solution is filtered, the chloroform recovered by distillation, and the residue in the flask dried. This residue consists of resinous fatty matters and theine ; the former are removed by treating the contents of the flask with hot water and 10 grains of powdered glass, which have been previously washed with dilute hydrochloric acid. The water is boiled and the contents shaken up with the glass ; the resinous matters attach themselves to the latter in the form of little globules. The solution is poured on a wet filter', and the residue completely exhausted by repeated boiling with fresh qiiantities of water. On evaporating the united filtrates in a tared capsule, pure cafi^eine is left in the form of white crystals. Dragendorfi" takes 5 grms. of the substance, exhausts it with boiling water, evaporates to dryness, adding 2 grms. of burnt magnesia and 5 of ground glass ; the finely powdered residue is soaked in GO cc. of ether for twenty-four hours, and finally thoroughly exhausted by ether. The latter, when separated and evaporated, leaves the theine in a tolerably pure state. He also states that ether may be replaced by clilorofbrm. Cazeneuve and Caillot recommend a very similar process, but magnesia is § 201.] THE CHEMICAL ANALYSIS OF TEA. 413 replaced by recently flaked lime,* ether by chloroform. Markow- nikoff uses benzine instead of the solvents mentioned. In all the above processes there is one source of error which does not appear sufficiently guarded against — viz., loss of theiue tluring the evaporation to dryness, since it is absolutely impossible to evaporate a decoction of tea and magnesia to dryness at 100° without loss of the alkaloid — a loss which, so far as the author's experiments go, does not take place until the mixture is quite dry. The following modification may therefore be proposed : — 4 to 5 grms. of the tea are boiled in a flask provided with an inverted Liebig's condenser for a couple of hours, the liquid and leaves are transferred to an evaporating dish, some magnesia added, and the whole concentrated to a pasty condition. This paste is treated and thoroughly exhausted by chloroform ; the latter is separated and evaporated, and the chloroformic extract redis- solved in a little boiling water, the solution filtered, evaporated to dryness at a very gentle heat, and weighed. (2.) Simple Treatment of the powdered Leaves hy Solveiits. — Legrif and Petit soften the leaves first with boiling water, and then extract the moist mass by the aid of chloroform. Other chemists simply exhaust the powdered substance by chloroform or ether ; subsequent purification may, of course, be necessaiy. (3.) SublimcUion. — A method of utilising tea dust by making it a source of theine, was recommended by Heijnsius (Journ. Frac. Chem., xlix. 317). The tea dust was simply treated in a Molir's benzoic acid subliming apparatus. Stenhouse improved this process by precipitating either a spirituous extract, or a decoction of tea by acetate of lead, evaporating the filtrate to dryness, mixing the residue with sand, and subliming. These processes of sublimation, however, were proposed simply for the extraction, not the estimation, of theine. The writer, in 1877,t proposed the following quantitative method of sublimation : — A convenient quantity of the tea was boiled in the way mentioned, magnesia added, and the whole evaporated to a paste, which was spread on a thin iron plate, and covered with a tared glass funnel. The heat at first was very gentle, but was ultimately raised at the later stages of the pro- cess to 200°. The theine sublimes perfectly pure and anhydrous, and forms a coherent white coating on the sides of the funnel ; the increase of weight is simply anhydrous theine. To ensure success it is absolutely necessary — (1.) That the layer be as thin as possible. * The present writer does not believe that magnesia can be replaced with lime without loss of theine from decomposition, t Op. cit. 414 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 202. (2.) That the heat be only gradually increased. (3.) That the mixture be occasionally cooled, and then thoroughly stirred. (4.) That the sublimation be prolonged for a sufficient time. The sublimation is finished when a funnel, inverted over the su.bstance, heated to about 150'', and left for half an hour, shows no crystals. An improvement on this process is to place the paste on a ground glass plate, to which a flanged funnel has been ground so as to tit air-tight. The apparatus is then connected with a Lane-Fox mercury-pump, and an absolute vacuum produced. By the aid of a shallow sand-bath, the theine may be sublimed at a very gentle heat. § 202. Determination of Total Nitrogen. — Peligot and Wanklyn have laid particular stress on the large amount of nitrogen contained in tea leaves. This nitrogen is, of course, lai'gely dependent on the theine, and it is questionable whether, with the improved methods for the extraction of the latter, it is worth while to make a combustion, more especially as the exhausted leaves are highly nitrogenous, from the presence of an albuminous body. The process is conducted in the usual way in a combustion tube, and best with coi:>per oxide. The following are a few determinations of total nitrogen : — Per cent. Analysed by A sample of genuine tea from Cacliar, . 4'74 Hodges. A hybrid variety, do., . 2 -81 ,, Another sample from Cachar, . . 4*42 ,, Sam]ile taken from 60 green teas slightlj' ) ^.^r. faced, \-^ '^ 60 Black teas, 3-26 Wirjner. 6 Assam teas, ..... 3 '64 ,, 6 Caper teas, 3 32 Assam tea, from Dr. M'Namara's garden, 3 '88 ,, Sami^le of exhausted leaves, . . . 3"80 ,, Mr. Wanklyn has applied his ammonia process to the examina- tion of tea. The soluble matter from 100 mgrms. of tea is heated with a 10 per cent, solution of potash in a flask fitted to a proper condenser, until all the ammonia is distilled ovei\ It may be necessary to add water once or twice, and redistil ; then 50 cc. of a strongly alkaline solution of permanganate of potash are added and distilled ; the ammonia in the distillates is estimated by " Nesslerising." Mr. Wanklyn gives the following figures as yielded by a genuine tea — MKi-tns. Free Ammonia, 0-28 Albuminoid Ammonia, 43 0-71 § 203.] THE CHEMICAL ANALYSIS OF TEA. 415 100 mgrms. of genuine tea, sent to the writer by Dr. Shortt, of Madras, yielded total ammonia -81 ] but this is a method which has not been accepted by chemists, although it has some value. § 203. Determination of Tannin. — The methods proposed for the determination of tannin are very numerous. Four only, however, require any notice here — viz., the gelatin process, the copper process, Mr. Allen's acetate of lead process, and Lowen- thal's process. (1.) By Gelatin. — The best process by gelatin is decidedly that which dispenses with the drying and Aveighing of the pre- cipitate. A solution of gelatin is carefully made by first soak- ing the gelatin in cold water for twelve hours, then raising the heat to 100°, by placing the bottle on the water-bath (the strength should be about three per cent.), and finally about -8 per cent, of alum should be added. A portion of the solution thus- prepared is put into an alkalimeter fiask {e.g., Schuster's), and carefully weighed. A solution containing a known quantity of tannin is now titrated with the gelatin until a precipitate no longer occurs; the flask is re weighed, and the loss shows approxi- mately the strength of the solution. One or two more exact determinations will be required to get the correct value. It is necessary to allow the precipitate now and then to settle, and a few drops of the supernatant fluid should be placed on a watch- glass, to which a drop of gelatin may be added, and thus the point of saturation ascertained. The tannin in a decoction of tea. is, of course, estimated on precisely similar principles. (2.) Copper Process. — When a single determination of tannin is required it is best to precipitate by coppei'-acetate. 2 grms. of tea are boiled for an hour in 100 cc. of water, the solution fil- tered, the filtrate boiled, and while boiling 20 to 30 cc. of solution of copper acetate [1 : 20] added. The precipitate is collected, dried, burnt to an ash, oxidised with nitric acid, and again ignited and weighed. 1 grm. of CuO = 1-3061 of tannin, if Eeler's* figures be accepted ; if Woolf's,t then 1 gnn. CuO = 1-304 tannin. (3.) Mr. Allen's Lead Process. — A filtered solution of lead acetate •5 per cent., a solution of 5 mgrms. of pot. ferridcyanide, 5 cc. of strong ammonia water, and 5 cc. of pure water, and lastly, solu- tion of pure tannin (-1 per cent.) are required. The process essentially depends upon the precipitation of tannin by lead acetate, and using ammoniacal pot. ferridcyanide as an indicator. The latter agent strikes a pink colour with tannin. The solution is standardised by taking a known volume of the lead solution, * Dingier' s Poly. Journ., 2-29, 81. •\ Zeitschrijtf. an. Chem., 1, 104. 416 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 203. and dropping in tlie tannin liquid until a small portion filtered gives a pink colour with the indicator. Tea is tested in a precisely similar manner. Mr. Allen's method is tolerably speedy and accurate ; the writer has, how- ever, found the final reaction somewhat difficult to observe. (4.) LowenthaVs Process. — Up to the present time this method (originally worked out for barks) is the best we possess ; it depends on the oxidation by permanganate, and indigo is used as an indicator. It not alone gives us the tannin, but the amount of other astringent matters as well. The following solutions are required : — (1.) A solution of potass, permanganate, 1*333 grms. per litre. (2.) Precipitated indigo, 5 grms. per litre. (3.) Dilute sulphuric acid (i : 3). (4.) A solution of gelatin, 25 grms. to litre, saturated with table- salt.* (5.) A saturated solution of pure salt, containing 25 cc. of sulphuric, or 50 cc. of hydrochloric acid per litre. The analysis as applied to the determination of tannin in barks is performed thus : — 10 grms., say, of sumach are taken and exhausted by boiling with water, and the solution made vip to 1 litre ; of this infusion, 10 cc. are mixed with 75 cc. of water, 25 cc. of the indigo solution added, and 10 cc. of the dilute sulphviric acid. The permanganate solution is run drop by droj) from the burette with constant stirring, till the blue colour changes to yellow, when the amount of permanganate used is noted {x). The same process is repeated with indigo and sul- phuric acid, and the amount read off" [y) ; subtracting y from X = total astringent matters. The permanganate oxidises both tannin and indigo ; but the tannin being the easier to oxidise, is consumed first. In order to obtain accurate results, the propor- tion of indigo should be such as to require about twice the quantity of permanganate which would be consumed by the tannin alone. Thus, if indigo alone requires 10 cc. of perman- ganate to decolorise it, the indigo and tannin together must not take more than about 15 cc. ; if it does so, the tannin must be diluted accordingly. The total astringent matters being known, the next step is to throw the tannin out, and estimate the gallic acid and impurities. 100 cc. of the infusion are mixed with 50 cc. of the salted gelatin infusion ; after stirring, 100 cc. of the salt acid solution are added, and the mixture allowed to stand * Lowenthal prepares the solution by steejiing 25 grms. of the finest Cologne glue in cold water over night ; it is then melted on the water-bath, saturated with NaCl, and made up to 1 litre with saturated KaCl sohition, filtered, and kept well corked. § 203.] THE CHEMICAL ANALYSIS OF TEA. 417 for twelve hoitrs. It is then filtered, and an aliquot part of the filtrate is oxidised by permanganate and indigo, as before. Lbwenthal gives the following example : 10 grms. of sumach were boiled in 750 cc, and after cooling made up to one litre : — Permanganate. (1.) 10 CC. of sumach infusion, 1 consumed 25 cc, of indigo solution, J > • • Do., repeated, 50 cc. of indigo solution alone, Total permanganate for 20 cc. of sumach, 50 cc. filtrate from the gelatin, \ „„j,s.„~„(i 25 cc. indigo solution, J ' ' Do. , repeated, 16-6 16-5 33-1 13-2 19-9 112 111 22-3 132 91 50 cc. indigo alone Gallic acid and impurities, .... Deducting 9-1 cc. from 19-9 cc. equals 10 '8 cc. as permanganate, equivalent to the tannin of 20 cc. of sumach infusion, or 0*2 grm. of dry sumach. It is well to ascertain the value of the perman- ganate solution by oxalic acid, adopting the numbers given by Neubauer and Oser — viz., that 0'063 oxalic acid is equal to 0*04157 gallo-tannic and -002355 quercitannic acids. Should it be pre- ferred to use tannin, the purest commercial tannin must be preci- pitated by lead, the precipitate freed from lead in the usual way, and the solution of pure tannin then evaporated to complete dry- ness, and a solution of convenient strength made. The process requires but little modification to be applicable to tea.* The amount of tannin in genuine teas seems to be variable, S. Janke, using the acetate of copper process, has determined the tannin in eighteen samples of black tea, and found as a maximum 9'142 per cent., as a minimum 6"922 per cent., and as a mean S'l per cent. Three samples of green tea gave 9*94, 8"56, and 9-57 per cent. Mr. Wigner, as a sample of very astringent teas, gives the following: — Per cent. Moyone young Hyson, . . . . . . 39'0 Very choice Assam, 33 '0 Indian young Hyson, . . , . . .39*0 Assam tea from Dr. M'Namara's garden, . . 27'7 Caper, mixed, 42'3 * F. Becker has proposed (Chem. Neivs, xli., 229) to estimate tannin as follows :— 50 cc. of a solution of methyl violet — "5 per cent, strength— is made up to half a litre and treated to 50°, it is then standardised by a 1 per cent, solution of tannin ; which is run in until all the colouring-matter is pre- cipitated and the filtrate is colourless. When operating on solutions of unknown strength, guided by the first rough essay, they should be diluted or concentrated to about 1 per cent. If this method be applied to tea, it would be well to prepare some pure gallo-tannic acid from tea, and to use this substance for the purpose of standardising. 418 FOODS : THEIR COMPOSITION AND ANALYSIS. 204. Exhausted tea leaves yield from 2 to 4 per cent, of tannin, A tea giving only 6 per cent, of tannin is to be regarded as suspicious, but care must be taken not to rely upon any single indication. § 204. The Extract. — The extract is a measure of the soluble matter in tea. Peligot exhausted the leaves and then redried them, and thus estimated the soluble matter by difference. Wanklyn, however, has proposed a more rapid and convenient method. It consists in taking 10 grms. of tea, and boiling with 500 cc. of watei-, the flask being adapted to a Liebig's condenser. When 50 cc. are distilled over, the process is stopped, and the 50 cc. retiirned to the flask ; 50-3 grms. of the hot strained liquid are then weighed out and evaporated to dryness. Wigner boils ■with a vertical condenser for an hoiir, and finds that 1 per cent, strength yields the most constant results. Perhaps, on the ■whole, the best process is the following : — Place one part of tea in 100 of water, boil for one hour with a vertical condenser, and then take an aliquot part of the filtered liquid for evaporation. In every case the time occupied in boiling, and the strength, should be mentioned in reporting, for two analysts operating by difierent methods may difier as much as 6 or 8 per cent. — the soluble matter not being entirely removed for a very long time. Since the substances that are at once dissolved are really those upon which its commercial value depends, it is a question whether it would not be better simply to pour boiling water on the leaves, let the infusion stand for one hour, and then estimate the exti'act, calling it extract of infusion. Any addition of exhavisted leaves lowers the percentage of extract. The following are some determinations of extract : — Per cent. Analysed'by Java tea, dried, 35-2 Peligot. ,, not dried, . 32-7 ,j Pekoe, ordinary, dry. 41-5 „ ,, undi'ied, 38-0 Gunpowder, dry, 51-9 ,, ,, uudried, 48-5 jj dry, . 46-9 „ „ undried, 50 '2 ^^ Moyone Gunpowder, 40-7 39-3 38-5 37-9 33-3 wYgner. Imperial, diy, . 43 1 Peligot. ,, not dried, . 39-6 „ dry, . . 47-9 11 not dried, . 44-0 >» § 205.] THE CHEMICAL ANALYSIS OF TEA, 419 Per cent. Analysed by. Hyson, dry, . 47-7 PeUgot. ,. not dried, 43-8 ,, Hyson skin, dry, . 43-5 ,, ,, not drie d, . 39 8 ,, Congou, . 36-8 »> ,, dried. 40-9 „ bon, . 40-7 ,, „ ,, dried, 45 jj " o3 Wigner. 29-8 „ '' [ 29-8 ,, '' 26-2 ^^ 261 Caper, dried, . 39-3 PeHgot. ,, not dried, 35-8 ,, 37-9 Wigner. 11 • '' * 37-7 ^^ 32-4 11 300 ,, Assam, dried, . 45-4 Peligot. ,, not dried, . 41-7 ^^ 33-3 Wigner. Hyson, 36-8 ,, Moyone Young Hys on, . 44-8 ,, Tea direct from Chii la, dry, 41-7 40-2 41-2 Wanklyn. Indian Tea, dry, 33-9 A.'Wynter Blyth, _ 43-S Wiguer. Broken Indian, 43-4 ,, Indian Souchoug, 32-5 11 Scented Orange Pek oe, . 34-2 ., Manuna, fiue, . 370 ^ Himalayan Tea, 386 35-4 Wanklyn. Since tlie extract of genuine tea appears to vary from 26 per cent, up to more than 40 per cent., it is unfortunately of no very great value for purposes of valuation. The extract, after being weighed, is burnt up to an ash, which will always be found to be heavy, rich in alkaline salts, and varying usually from 4 to 7 per cent. § 205. The Ash. — The percentage of total ash is taken by- burning up 1 to 5 grms. of the tea in a platinum dish. The leaves readily ignite, and the operation may take place at a very- low temperature, so that there is, with care, very little volati- lisation of chlorides. The comparative composition of the ash of fresh and of exhausted tea leaves is shown in the following table: — 420 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 205. TABLE XXIII. ZOLLEB. Hodges. ZOLLEB. "WiGNEK. Ash of fine young Himalaya Tea. T«a from Caohar (indigen- ous). Tea from Caehar (hybrid). Exhausted Tea Leaves. Ash of a number of Mixed Black Teas Ash of a number of Mixed GreenTeas Potash, . 39-22 35-200 37-010 7-34 30-92 28-42 Soda, _ . 0-65 4-328 14-435 0-59 1-88 2-08 Magnesia, . 6-47 4-396 5-910 11-45 ... Lime, 4-24 8-986 5-530 10-76 Oxide of Iron, . 4-38 2-493 2-463 9-63 Manganous Oxide, 1-03 1-024 0-800 1-97 ... ... Phosphoric Acid, 14-55 18-030 9-180 25-41 Sulphuric Acid, . trace. 5-040 6-322 trace. 4-SS 5-66 Chlorine, . 0-81 3-513 2-620 trace. Silica and Sand, 4-35 0-500 1-300 7-57 1-70 7-50 Charcoal, . 2-900 1-830 Carbonic Acid, . 24-30 13-590 12-600 25-28 11-60 6-43 100-00 100-00 100-00 100-00 Percentage of total Ash sol- [ 57-00 52-85 uble in ■water. Tlie ash, on being cooled and weighed, is next boiled up with a little water, the soluble portion filtered from the insoluble, and washed in the ordinary way. The filtrate is evaporated to dry- ness, very gently ignited, and returned in percentage as soluble ash. The insoluble portion is next treated with acid, and the remaining sand dried, ignited, and weighed. The alkalinity of the soluble portion should also be taken, and may be returned as potash. This simple examination of the ash, consviming very little time, gives tolerably well all the information afibrded by a complete and exhaustive analysis. The table (XXIV.) shows a few percentages of ash, and may be compared with the per- centages of the ashes of beech, bramble, and others. All the analyses hitherto pv;blished show that the percentage of ash in genuine tea never reaches 8 per cent. An ash beyond 8 per cent., calculated on the dried tea, is certainly adulterated. In the same manner, all genuine tea possesses a soluble ash not § 205.] THE CHEMICAL ANALYSIS OF TEA. TABLE XXIV. 421 «> . s ^^ S'l- 3--- u m 2^U 1 •s Authority. g« p£ r^ M S Average of 17 ordinary \ Teas from original | chest, consisting of 2 I Indian, 12 Congous, i 5-75 3-07 2-25 0-43 1-38 G. W. Wigner. 2 Gunpowders, and 1 1 Hyson, . . ) Maximum, 603 3-35 2-87 0-76 1-88 J Minimum, 5-53 2-75 1-99 015 1-17 Average of 25 special ) Teas, . . \ 5-95 3-33 209 0-53 1-38 ,, Maximum, 7 02 3-88 2-68 1-67 1-96 Minimum, 517 264 1-33 0-04 1-08 '^ Genuine Indian Tea, . 5-61 2-90 A.WynterBlyth. Common Tea, 5-92 3-55 Wanklyn. Paraguay Tea, . 6-2S 4-22 ..'. ^, Average of 7 Teas, 5-75 A. S. Wilson. 9 „ 5-66 s-oo ... A. H. Allen. Horniman's p. black, 5-30 3-50 ... green. 5-60 3-80 ... ... ]' Ambrosial black. 5-60 3-40 Genuineblk.,2s.6d.lb., 5-60 5-70 6-02 6-34 6-10 5-75 309 3-28 3-26 3 20 3-96 3 06 ... • " 3s."lb., . 5-50 3-55 Broken leaf, with stalks, 5-40 2-80 " Caper (4-S silica). 11-40 1-50 ... " " Mixed dry exhausted ) leaves from various > 4-30 0-52 Teas, . . \ " Coffee leaves. 10-32 3-77 Beech, 4-52 200 Wanklyn. Bramble, , 4-53 1-84 ... ... Raspberry, 7-84 1-72 ... '' Hawthorn, 8-05 3-78 " Willow, . 9-34 4-16 ... " Plum, 990 5-66 ..] ... " Elder, 10-67 319 ... " Gooseberry, 13-50 7-83 '"' less than 3 per cent. For examples of obviously impure ashes, Mr. Wigner's paper may be quoted from again : — 422 FOODS : THEIR COMPOSITION AND ANALYSIS. [§207 Total Ash. Ash Soluble in Water. Soluble in Acid. Silica. Alkali calculated as Potash. Extract. Gunpowder, . Caper, . 19-73 14-44 15-20 15-08 12-74 14-60 1-00 1-95 169 1-96 2-68 2-67 6-15 2-47 5-35 5-65 5-44 5-67 12-58 10-02 8-16 7-47 6-62 6 06 0-14 103 0-61 0-73 1-04 1-04 37-78 35-45 31-60 35-60 All these teas, although imported in this state, are evidently mixed with sand to a considerable extent. § 206. Determination of Gum. — If it is necessary to determine the gum in tea, as sometimes hapjsens, the aqueoiis decoction should be evapoi-ated nearly to an extract, and the residue ti'eated with methylated spirit, filtered, and washed with the spirit. The gum is dissolved oft" the filter by the aid of hot water, and the solution evaporated to dryness, and weighed ; it is then ignited to an ash, and the mineral deducted from the total weight.* § 207. General Eeview of the Adulterations of Tea. — The most frequent are, certainly, the addition of sand, generally strongly impregnated with iron, the addition of foreign and exhausted leaves, and the addition of astringent principles, such as catechu, &c. All these adulterations must take place abroad, there being no evidence that a single hundredweight of tea has been tampered with in England, — the blame may lie with the home-traders, but proof is wanting. On the other hand, it not unfrequently happens that cargoes of tea recovered from sunk vessels, or teas damaged in some other way, are sold and blended by wholesale manufac- turers with those that are genuine. Such samples contain usually * H. Hager is (PJiarm. Central. Halle, 1879, 258) the author of a general process of analysis, which possesses some good points : 10 grms. of tea are infused in 100 cc. of warm water, and allowed to stand for two days ; the sokition is poured off, and another 100 cc. of water added to the partially- exhausted leaves, which are then unrolled and botanically examined. 50 cc. of the solution are evaporated to dryness ; 10 cc. of the solution should give no turbidity in the cold when an equal volume of alcohol is added. For the estimation of theine, 10 grms. of tea, 3 grms. of sodic car- bonate, and 3 grms. of litharge are made into a paste with 10 grms. of water dried up and extracted with chloroform. For the special detection of catechu, -1 grm. of tea is extracted by 100 cc. of boiling water. This solu- tion is boiled with excess of lead oxide, and the filtrate (which must be clear) mixed with a solution of silver nitrate. Pure tea gives only a slight grey-black depr)sit of metallic silver, but tea adulterated with catechu a strong yellow flocculent precipitate. § 208.] THE CHEMICAL ANALYSIS OF TEA. 423 an excess of salt, and show more or less evidence of the addition, of exhausted leaves. The facing of tea is rapidly decreasing. There has been much dispute as to whether this is to be considered an adulteration or not ; a tliin film of gi^aphite, or any other harmless substance, in such quantity as to add no appreciable weight, can hai-dly be called adulteration. Each case, however, must be judged of by its merits. A small addition of such a substance as catechu, to im- part astringency, is probably frequent, and difficult of detection. Any amount present, to the extent of 3 per cent, or over, is shown by precipitating an infusion of the tea with a slight excess of neutral lead acetate, filtering, and adding a little dilute ferric chloride solution. If catechxi be present there is a bright-green colour, and ultimately a precipitate of a gi^eyish-green colour. [See also Hager's method, footnote, p. 422.] The same infusion filtered from the lead precipitate gives a copious precipitate with argentic nitrate. Mr. Allen has pointed out the advantage of his lead process in cases of adulteration with catechu, and it is self-evident ; for catechuic acids possess a precipitating power so widely difi"erent from that of tannin, that, if reckoned as tannin, there are always anomalous results, indicating a much higher astringency than could possibly exist, — e.g., a sample of brown catechu examined in this way, and reckoned as tannin, gives the paradoxical number of 11 per cent. Soluble iron salts, alkaline carbonates, and other substances, are stated to be occasionally added, but no conviction relative to these appears to be on record. The soluble iron salts may, of course, be dissolved from the tea leaves by a little cold dilute acetic acid, and the liquid tested in the usual way; there is then no confusion between the iron naturally present and that added. § 208. Bohemian Tea.''' — It woiild seem that for some time there has been cultivated in Bohemia the Lithospermum officinale, the common " Gromwell" of our country, and the leaves have been dried and sold as Thea Chinensis, under the name of " Bohe- mian Tea." They have also been used for the purpose of adulterating Chinese tea. The " G-romwell " is a plant belonging to the borage order, growing in dry and stony places, from a foot to a foot and a half high. The flower is greenish-yellow, the stem erect and branching, and the leaves are lanceolate, hairy beneath, with bulbous adpressed bristles above. They are totally unlike tea leaves, and the hairiness itself would be diagnostic of a leaf other than that of tea. The chemical composition is also entirely difi'erent. The mineral constituents are excessive, and there is neither any alkaloid nor any essential oil. The average composition of " Bohemian Tea " is as follows : — * A. Belohouben: Chem. Centralhl, 18S0, p. 152. 424 FOODS : their composition and analysis. [§ 208. Cellulose, 5-9637 Tannin, 8-2547 Fat, 9-2910 Other nitrogen-free organic substances, . . . 26-4941 Albuminous matters, 24-5406 Ash, 20-5960 Water, 4-8599 BIBLIOGRAPHY EELATIVE TO TEA AND THEINE. Allen, A. H. — Pliarm. Journ., 1873; Ghem. Nevis, vol. xxx., 1874. Ball. — "Account of the cultivation and Manufacture of Tea in China." Becker, F. — Estim. of Tannin, Chem. News, xli., 229. Berthemot u. Dechastelus— /oM?-7i. Pliarm. (2), xxvj. 63. Blyth, a. W. — Detect, of Foreign Leaves in Tea, Analyst, June, 1877j Indian Tea, Journ. Chem. /S'oc, 1875, 385. BoLLEY — Journ. Fract. Chem., xciv. 65. Brown, Dr. Campbell. — Journal Chem. Soc, 1875, 1217. Cazeneuve, Paul, and 0. Caillot. — Bull. Soc. Chim. [2], xxvij. 199. Commaille.— JoM?-tt. de Pharm. d'Anvers, 1876, 121. Davis. — Journ. Pharm. et Chim., 3e se'rie, 1853, t. xxiv., p. 228. Fortune, Robert.— i?e^er<. de Pharm., 1856, t. vij., p. 117. Garot.— reci])itating with sugar of lead, retiltering, freeing the filtrate from excess of lead by iSHj, evaporating to dryness, and subsequent purifying of the * 20 per cent, is the standard of the Society of Analysts; but this, according to published analyses, is much too low. § 222.] COCOA AND CHOCOLATE, 453 residue by solution in spirit, and treatment with animal charcoal. Mits- cherlich, again, boils the cocoa with a weak solution of sulphuric acid iu order to change the starch into sugar, saturates the fluid with carbonate of lead, and ferments it with yeast to destroy the sugar. On the couclusioa of the fermentation, the fluid is boiled, neutralised with soda. Altered, concentrated by evaporation, and the impure brown theobromin which separates boiled in hot nitric acid. This nitric acid solution is precipitated by ammonia, again dissolved in nitric acid, and the nitrate obtained by evaporation. According to Mitscherlich, the quantity obtained in this way is much greater than by other processes. Wm. E. Kunze* has made an exhaustive study of the pub- lished processes for the separation and estimation of theobromin and caffein in cocoa of Weigmann, Mulder, Wolfram, Legler, Trojanowsky, Zipperer, Siiss, Diesing, and James Bell, and finds them all imperfect or inaccurate ; and after much research, he has adopted the following, which, from test analyses, appears to give results of great accuracy : — 10 grms. of cocoa are boiled with 150 cc. of 5 per cent, sul- phuric acid for 20 minutes, the fluid is filtered, and the residue well washed with boiling water. While warm the mixture is precipitated by phosphomolybdic acid,t and after standing 24 hours the precipitate is filtered and washed with much 5 per cent, sulphuric acid. The precipitate while still moist is transferred to a flask, and decomposed with baryta water. The alkaline fluid is then saturated with CO., and evaporated to dryness. The dry mass is transferred to a flask, and exhausted by boiling chloroform, the flask for this purpose being connected with an inverted condenser. The chloroform extract is placed in a tared flask, the chloroform distilled ofl^, and the residue dried and weighed. The weight equals theobromin + cafleiu, with a little mineral matter, so small that, in technical analyses, it may be neglected. If it is desired to estimate these crystalline substances separately, the residue after w^eighing is dissolved in a little water, ammonia added, and to the ammoniacal solution silver nitrate added. On prolonged boiling the crystalline silver theobromin falls down (C^HyAgN'^Oo) containing 37*488 per cent. Ag. This may be weighed, or dissolved in dilute nitric * Zeit. f. analyt. Chem., 1894, 1. See also Brumner and Heinrich Leins, Schueiz. Wochen. f. Chem. u. Pharm., March, IS9.:J. t Prepared by precipitating ammonium niolybdate, by sodic phosphate, suspending the well-washed precipitate in water, and adding sodium carbonate, and warming until dissolved. The solution is evaporated to dryness and ignited. The ignited residue is powdered, mixed with nitric acid into a paste, and again ignited. The residue is warmed with water, nitric acid added to strong acid reaction, and diluted to 10 per cent. ; the solution is warmed for some time and ultimately filtered ; the result should be a clear yellow liquid. 454 FOODS : their composition and analysis. [§ 222. acid, and the silver precipitated as chloride. Ivunze, however, prefers a volumetric process of separation. He adds a measured quantity of silver nitrate solution of known strength (say 5 per cent.), and after filtration, titrates the unused silver by ammonium sulphocyanide — by multiplying the silver tlius found by difference in the precipitate by the factor 1-66 (for 108 silver = 180 tlieobromin) the weight of the theobromin is obtained, and this weight subtracted Irom the combined weight of theobromin and cafl'ein gives, of course, the caflein. Theobromin forms microscopic rhombic needles. It is gene- rally thought to sublime between 296° and 295° without decomposition, but this temperature is many degrees too hi^h. The writer finds that a minute fragment, placed in the sub- liming cell elsewhere described, begins to give fine nebulse at 134'', and on examining the mists by a high power, they ai'e resolved into extremely minute dots ; distinct crystals ai-e obtained at temperatures of 170° and above. Theobromin is insoluble in petroleum ether, and not very soluble in ether, 1 part requiring 600 parts of boiling and 1,700 parts of cold efcher. It is soluble in alcohol, 1 part requiring 47 parts of boiling and 1,460 of cold alcohol. Its solubility in water is stated to be 1 in 55 parts at 100°, 1 in 600 parts at 20°, and 1 in 1,600 at 0°. It is somewhat soluble in chloroform and ■warm amyl-alcohol, but with difficulty soluble in benzole. Theobromin* forms easily crystallisable salts. The simple neutral salts are decomposed by water, with the formation of basic salts, and lose their acid, if it is volatile;, at 100°. A hydrochloride of theobromin, O^HglSr^O.^jHCl ; a nitrate, CyHsN^Og.NHOg; a platinum salt, C7H8N40o,HClPtCl2 + 2H2O; are all very definite crystalline compounds. A noteworthy salt is the nitrate of silver, which is formed by adding a solution of argentic nitrate to a solution of nitrate of theobromin; in a short time there sepai'ate silver-white needles, very insoluble in water, of the composition C7HsN^O._,,NH03 + AgNOo. The other preci[)itants of theobromin are — phosplio-molybdic acid (yellow) and chloride of gold (long needles). Tannic and picric acids only produce turbidity, while potass, mercuric iodide and potass, cadmium iodide do not precii)itate. A characteristic reaction of theobromin is that produced by peroxide of lead and sulphuric acid. If peroxide of lead and diluted sulphuric acid * E. Fisc-her {Ber. xv., 453-456) has transformed xanthine, C5H4N4O2, into theobromin by heating- lead xauthinate, dried at 130° in a sealed tube, at 100° with i its weight of methyl iodide. § 222a.] COCOA AND CnOCOLATE. 455 are heated with theobromin, avoidin!:;; an excess of the oxidising agent, CO., is developed, and if filtered from lead sulphate the filtrate gives ofi" ammonia with potash, separates sulphur on treatment with SHo, stains the skin purple-red, and colours magnesia indigo-blue. Theobromin is poisonous to kittens (and other animals of similar size) in sucla large doses as a gramme. It appears to be separated by the kidneys, and coukl probably be discovered in the urine of any person taking large quantities of cocoa. The method of research successfully used by Mitscherlicli is as follows : — The urine is acidified with HCl, filtered, and to the filtrate, acidified with nitric acid, a solution of phospho molybdate of soda is added. The precipitate is collected, and treated with baryta water until it is strongly alkaline, warmed, filtered, and the filtrate evaporated; the residue extracted with alcohok refiltered, and the filtrate again evaporated. This last residue is dissolved in a drop of hydrochloric acid, and precipitated by ammonia. The alkaloid may now be collected and, if necessary, purified. § 222a. Cocoa red is an astringent colouring-matter found in cocoa. On saponification it breaks up into glucose, tannin, resin, and the ill-defined brown amorphous substance to which has been given the name of phlobapheue. The author has been working at cocoa red lately, and so far has ascertained the following facts : — When cocoa is freed from fat the red colouring-matter is only partially extracted by sol- vents, unless a mineral acid has been added. By adding a few cubic centimetres of hydrochloric acid and gently warming for a few seconds, the red colouring-matter is dissolved with great ease by amyl alcohol, as well as by ethyl alcohol. It can be in part shaken out of an aqueous solution by amyl alcohol. It is insolu- ble in ether or petroleum ether, but dissolves to a slight extent in carbon disulpbide. The be.st way to extract cocoa red is, apparently, as follows : — From 2 to 3 grms. of the fat-free cocoa are made into a paste with hydrochloric acid, the acid paste put into a Soxblet and exhausted by 100 cc. of absolute alcohol, the alcohol being heated in a flask standing in a beaker of boiling water. Before placing the alcohol in the flask, sufficient silver oxide is added to the alcohol in the flask to fix the hydrochloric acid. The alcoholic liquid is cooled, filtered, and then precipitated by an alcoholic solution of lead acetate. The precipitate is of a purple black. It is collected on a filter, well washed with boil- ing water, and then transferred to a small flask, some 70 per cent, alcohol added and the lead salt decomposed by SH.,; on getting rid of the SHg, filtering and evaporating to dryness, the red colouring-matter is obtained in a solid form, and by repetitioa of the process purification may be efiected. 456 FOODS : their composition and analysis. [§ 222a. A solution of cocoa red obtained in this way gives a diffuse band in the green, allowing the red, blue, and most of the yellow rays to be transmitted. The solution in alcohol is capable of being estimated on colorimetric principles, but low results are obtained because the process always entails some decomposition. If estimated by quantitative spectroscopy, according to the author's observations the wave lengths between 590 and 578, between 578 and 565, and between 539 and 529 are the best adapted. The absorption factors for these wave lengths are as follows: — X 590 to 578, 0-001332 \578 ,,565 0-00148 K539 ,,529, 0-000995 Cocoa red is a sensitive reagent to acids and alkalies, alkalies generally striking a dirty green, mineral acids a red with a violet shimmei'. The solution in water is bitter, and gives precipitates with the salts of iron, copper, and silver. Zipperers Method of Determinmg Cocoa Red and the Products of its Deco7nposition. — 100 grms. of cocoa are exhausted of fat by petroleum ether. The residue is dried and then macerated for 8 days in a litre of absolute alcohol. The alcohol is filtered off, and most of the alcohol is recovered by distillation ; in any case the alcohol is evaporated to an extract and weighed. This mainly consists of cocoa red with resin and phlobaphene, products of its decomposition ; the weight may be designated by p. The cocoa which has been exhausted by petroleum ether and treated with alcohol is now macerated in a litre of water for two days, at the end of which period it is filtered. The filtered liquid is precipitated by four times its volume of absolute alcohol and again filtered. The filtrate is evaporated to an extract. The extract is treated with a definite quantity of water, and a fractional portion of this is precipitated with neutral acetate of copper ; the tannate of copper is collected, dried, weighed, and incinerated, the loss of weight on incineration being returned as tannin; let this weight be denoted as p^; then p + p^ equals "cocoa red." The phlobaphene and resin may also be determined. To do this the alcoholic extract, p, after being weighed, is treated on a filter with acidulated water (which dissolves the tannin) and washed with water. The residue thus freed from tannin is washed with 200 cc. of ammoniacal water (50 per cent.) which dissolves the phlobaphene; and the solution is filtered, evaporated to dryness, and weighed as phlobapheno. §§ 2225, 223.] cocoa and chocolate. 457 The residue freed from tannin and phlobaphene is dried and weighed and returned as resin. Zipperer obtained from a sample treated in this way, p = 2-64 per cent., /^ =2-85, phlo- baphene 2-0, resin 0-07 ; in other words, the total cocoa red was 5-49, the tannin glucoside being equal to the difference between the weight of the total cocoa red and the united weight of the phlobaphene and resin — that is to say, 5 '49 - 2 '07 = 3-42 per cent. Zipperer's method with its large quantities of absolute alcohol is expensive and time-consuming, and without a doubt could be simplified. With careful work certainly a tenth of the quantities may be used — that is to say, 10 grras. of cocoa instead of 100, and the solvent diminished in the same proportion. § 2226. Determination of Crude Fibre. — A gramme of the substance is freed from fat by exhaustion in a Soxhlet with ether or petroleum ether. It is then boiled under an inverted condenser with 1-25 per cent, of dilute sulphuric acid; for this purpose from 100 to 150 cc. of acid will be required;, it is then filtered and thoroughly washed with hot water. The matter on the filter is next washed into a flask and boiled with 100 cc. of 1-25 per cent, soda solution. This is filtered and the residue thoroughly washed with hot water and, lastly, with alcohol, collected on to a platinum dish, dried, weighed, and ignited. § 223. The Ash. — The composition of the ash of cocoa seeds i& stated by Mr. Wanklyn to be as follows : — CoMPOSiTiox OF Ash of Cocoa Seeds. Per cent. Potash, 29 81 Chloride of sodium, G 10 Ferrous oxide. 1 60 Alumina, 2 40 Lime, . 7 7 "2 Magnesia, . 7 90 Phosphoric acid, 24 2S Sulphuric acid, 1 92 Carbonic acid. 9S Silica, . 5 00 Sand, . 12 15 458 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 223. The percentages of ash found in cocoa are given as follows : — Percentage of Ash. Common Trinidad, ...... 3"27 Very fine Trinidad 3 '62 Fair, good, tine Trinidad, ..... 3"64 Fine Grenada, 3-06 Caraccas, ........ 4"58 Bahia (Brazil), 3-31 Fine Surinam (small), 3-06 Mexican, 4-27 Dominican, 2-82 African, 2 "68 Mean of the ten being ..... 3"43 Thus the lowest determination is 2*68, the highest 4-58 per cent. The nibs show a lower ash than the shelh The nibs of the Caraccas give 3-95 per cent, of asli, 2-00 being soluble and 1-95 insoluble in water. The nibs of Mexican seeds give 2*59 per cent, of ash, -89 part being soluble and IwO insoluble in water. The ash of the shell is rich in, but the nib almost devoid of, carbonates. The ash should in all cases be dissolved in a measured quantity of d. n. sulphuric acid, boiled to expel carbonic acid gas, and then, after adding litmus, titrated back with d. n. soda ; the difference between the two results is a measure of the alkalinity of the ash, and is called by the American chemists the "acid equivalent." Dutch cocoas and others which have been treated with alkalies show a high alkalinity. Mr. Heisch has examined the cocoas of commerce, with the results embodied in the following tables : — TABLE XXIX.— Examination of Roasted Beans after removal OF THE Husk. "O a •a 4 < f^' j3 g 3 2 .Ho a p Oo- to g .%: &^ o 'o's 1 ^ is § ja -d ■?n o fe? a fa ■^ < lb ■* -* cot)< 1 tj< II 1 " O C5 — 1 COM ^ Ttt T)< in ITS lO ITS 1 CO ^ So p p ip CI ^ O^ CI CO C) O C) O c3 CI Id 1 d COCOM ■* -H CJ M CI lO O CO CO lO p C, ^ IQ ^ ] p ■? Air-dried husked beans— Maracaibo, Caraccas, . Trinidad, . Machala (Guayaquil), Portoplata, Means, Air-dried husks— Maracaibo, Caraccas, . Trinidad, . Machala (Guayaquii), Portoplata, Chocolate in cakes mixed sugar only — 4-8 marks per kilo., . 4'0 marks per kilo., . 3-2 marks per kilo., . 2-4 marks per kilo., . 2 marks per kilo., . Means, 462: FOODS : THEIR COMPOSITION AND ANALYSIS. L§ 223. TABLE XXXc— Numerical Relations between the various Con- stituents OF the Cocoas in the preceding Table as calculatei> BY Bexsemann. S = starch ; F = fat ; U = total organic matter insoluble in water. s F s tr F tr S F Air-dried husked beans — Maracaibo, . 0-42S9 0-7395 0-1636 0-6185 0-2645 Caraccas, . 0-4074 0-7273 0-1578 0-6125 0-2577 Trinidad, . 0-3452 0-7-297 0-1247 0-6387 0-1953 Machala (Guayaquil), . 0-3o83 0-73.30 0-1297 0-6379 0-2033 Portoplata, . Means, Air-dried husks — 0'3660 0-7734 0-1481 0-6589 0-2247 0-3946 0-7406 0-1446 0-6335 0-2283 Maracaibo, . 0-1390 0-0412 0-1341 0357 3-7564 Caraccas, 0-1525 0-0356 0-1479 0304 4-8674 Trinidad, . 0-1515 0-0467 0-1455 0-0399 3-6413 Machala (Guayaquil), . 0-1272 0402 0-1227 0-0.352 3-4827 Portoplata, . Means, Chocolate in cakes mixed with sugar only— 0-lSdO 0-0797 0-1728 0-06G0 2-6202 0-1508 0-04S4 0-1446 0-0414 3-4920 4-8 marks per kilo., . 0-3838 0-7303 0-143S O-G-253 0-2,300 4 marks per kilo., . 0-3476 0-7161 0-1314 0-62-20 0-2112 3-2 marks per kilo., . 0-3310 0-7226 0-1207 0-6354 0-1899 2-4 marks per kilo., . 0-3029 0-7035 0-1141 0-6232 0-1831 2-0 marks per kilo., . Means, 0-3729 0-7490 0-1298 0-6517 0-1992 0-3480 0-7245 0-1-282 0-6317 0-2029 COCOA AND CHOCOLATE. 465 §§ 224, 225.] § 224. Stutzer classifies the nitrogenous constituents of cocoa as follows : — 1. Non-proteids, substances soluble in neutral watei- solution in presence of copper hydrate (theobromin, ammonia, and amido compounds). 2. Digestible albumen. 3. Insoluble and indigestible nitrogenous substance. In three samples he found the mean numbers in per cent, of total nitrogen to be— soluble non-proteid nitrogen 29-4 per cent., albumen 32-2 per cent., and not digestible 38-4 per cent. Mr. Wigner's examination of the nitrogenous constituents of cocoa, based on 84 coininercial samples, are as follows : — TABLE XXXI. — Nitrogenous Constituents of Cocoa. i ^. ■§2-5 .2 ^3 ^3 ■ ■3 |3 a ■3 a'io ^2^^ .4?,^?,^'^ t^ ^ 1 If < " IP 1. 1-095 •600 9-92 3-80 3-12 54 9 2. 1-162 •760 7-35 481 2-54 65 5 3. 2-978 2-3.35 18-84 14-79 4 05 78-5 4. •965 •375 6 11 2-;^7 3 74 38 •S 5. •699 -.330 4-42 2-09 2-33 47^3 6. r201 -770 7-61 4-88 2-73 34^1 ( Socunza, . 2 -040 1-175 12-92 7-44 4-48 57-6 Roasted beans, ) Para, . 2-ooa 1045 12-67 6-62 6 05 52-2 1 Trinidad, , 1-490 1-050 9-46 6-65 2^81 70^3 ( Grenada, . 2-370 1-335 14-99 7-56 7-43 50-4 § 225. Adulterations of Cocoa. — The list of adulterations usually given is as follows : — Sugar, starches, Venetian red, brick-dust, and peroxide of iron. Some of these sophistications, svich as the starches, may be detected by a preliminaiy microsco])ical exami- nation, which in no instance should be neglected. The ordinary chemical examination consists in the extraction of the fat as. before described, the estimation of the percentage of ash in the ordinary way, its division into soluble and insoluble, and its content of phosphoric acid. By a simple estimation of the fat and the chief constituents of the ash, supplemented by the use of the mici'oscope, all known adulterations can be detected. The amount of phosphoric acid in the ash of soluble cocoas has. been taken as a basis of calculation of the amount of cocoa, and in the absence of foreign seeds, or other phosphate-producing 464 FOODS : their composition and analysis. [§ 226. eubstance, the calculation will be a fair approximation to the truth. The ash itself and the amount of phosphoric acid will, of course, be very notably diminished in the case of the soluble cocoas, and the percentage of the phosphoric acid will in such instances be a fair guide to the amoxmt of foreign admixture. For example, suppose a soluble cocoa to yield an ash of 1-5 per ■cent, -6 of which is due to phosphoric acid, taking as a basis of -calculation -9 per cent.* of phosphoric acid in cocoa nibs : — That is, the mixture contains about 66-6 per cent, of cocoa. The amount of starch in cocoa may be determined in the ordinary way, as described at p. 165 et seq., but the process is somewhat tedious, and may be dispensed with, since the extract in cold water is always a guide to the adulteration by starchy substances. Cocoa nibs treated in this way give to water about 6-76 per cent, of organic matter and 2-16 of ash. The determination of theobromin and of cocoa red is also of use. § 226. Achdterations of Chocolate. — Oil of almonds, cocoa-oil, beef and mutton fat, starches, cinnabar, chalk, and various other substances are usually enumerated ; a few of these are, however, apocryphal. The analysis of chocolate is conducted on exactly the same principles as that of soluble cocoa. If it is desired to separate the different constituents, the method recommended by A. Porrier may be used :• — Extract the fat with ether, and the sugar with alcohol of 20°, and dissolve the starch out by boiling water. The liquid holding the starch is then decolorised by animal charcoal, and the starch precipitated by alcohol of 50 per cent., dried, and weighed. But as regards adulteration, the procedure recommended in the case of cocoa will be found quite efficient, and less cumbersome and tedious. BIBLIOGRAPHY. Babbet. — Proc^d§ pour reconnaitre la quantite de f^cule contenne dans les chocolats. Journ. Chim. Mdd., 4e s^rie, 1S57, t. iii., p. 121. Bensemann. — Various papers on Cocoa and Chocolate. Ber. d. chem. Ges., xvi., 856; Eep. f. anal. Chem., iv., 213; Zeit. /. anal. Chem., xxiv., 628. * The lowest percentage given by Mr. Heisch, and but little different from Mr. Wanklyn's. § 226.] BIBLIOGRAPHY. 465 Bjorklund. — Moyen de reconnaitre la falsification du beurre de cacao, &c. Journ. Chim. Med., 5e s^rie, 1869, t. v., p. 146. BoussiNGAULT. — Various papers. Compt. Bend., xcvi., 1395; Ann. Chim. Phys., [5] xxviii., 433; Ann. Chem. Pharm., xxi. 200. Briois. — Essai sur la recherche et le dosage de la f^cule dans le chocolat. Journ. Chim. Med., 4e serie, 1856, t. ii., p. 497. Chevalliek. — Memoire sur le chocolat, sa prt^paration, ses usages, les falsifications qu'on lui fait subir, les moyens de les reconnaitre. Ann. d'Hyrj. et de 31id. Leg., 2e se'rie, 1844, t. xxxvii., p. 241 ; ib.. Journal d'Hg.,2S,lS16. Cooley's " Dictionary of Practical Receipts." Arts., Chocolate and Cocoa. EwELL, E. Erwin. — Cocoa Preparations in U. S. Dep. of Agriculture (Chem. Div.) Bull., No. 13, 1892. Glasson. — An7i. Chem. Pharm., Ix. 335. GiRARDiN et BiDARD.— Sur la fecule de cacao. Journ. Pharm. et Chim., 3e serie, 1852, t. xxxvij., p. 266. Heisch. — On the Composition of DiS'erent Kinds of Cocoa. Analyst, i., 1877, p. 142. HiRSCH. — "Die Priifung der Arzneimittel." HusEMANN. — "Die Pfianzenstofi"e. " Keller. — Ann. Chem. Pharm., Ix. 335. KiKGZETT, C. T. — The Chemistry of Cocoa Butter. Journ. Chem. Soc,, 1878, p. 42. Letellier. — Falsification des chocolats. Journ. de Pharm. et Chim., 3e serie, 1854, t. xxv., p. 362. Mansfield, M.—Bev. internationale et popul. des falsifications, iv. 40. MiTscHERLiCH, A. — " Dcr Cacao und die Chocolade." 1859. Muter, John. — On "Prepared" Cocoa. Analyst, 1879, p. 65. PoRRiER. — Sur le dosage des matieres amylacees ajoute'es frauduleusement au chocolat. Journ. Chim. Med., 4e serie, 1858, t. iv., p. 297. Strecker. — .47i7i. Chem. Pharm., cxviij. 151. •Stutzer. — Various papers. Bep. f. anal. Chem., 1882; Chem. Centrhl., 1882 ; Buss. Zeit. f. Pharm., xxi. 724. Traub, M. C— On Cocoa Butter. Arch. Pharm. [3], xxi. 19-23. Wanklyn. — "Tea, Coffee, and Cocoa." Lond. 1876. Wigner, G. W. — On the jSiitrogenous Constituents of Cocoa. Analyst, 1879, p. 8. Wolfram, G.—Zeitschri/tf. anal. Chem., 1879, s. 346. WosKRESENSKY.— ^7171. Chem. Pharm., xli. 125. .Zipperer, in KOnig's Chemie der mensch, Nahrungs- u. Genussmittel, 3 Aufl., Bd. 1. 13 PART VI.-ALCOHOL, SPIRITS, LIQUEURS. ALCOHOL. § 227. The term alcohol, in its strict chemical sense, applies only to the neutral compounds of oxygen, carbon, and hydrogen, which, by the action of acids, form ethers. The principal alcohols are enumerated in the following table : — ■ TABLE XXXIL, EXHIBITING THE PROPERTIES OF THE PRINCIPAL ALCOHOLS. Specific Gravity, Vapour. Boiling Point. Alcohols. Formula. Eel. Wt. Liquid. Vapour. H = l. Fahr. Cent. 1. Wood Spirit, or ] Methylic Alcohol, ( CH4O 0-798 1-12 16 149-9 65-5 2. Spirit of AViiie, or Ethylic Alcohol, CsHgO 0-7938 1-6133 =3 1730 78-3 3. Tritylic or Pro- pylic, . . .\ 4. Tetrylic or Butylic, C3H8O 0-817 2-02 30 206 96-7 C4H10O 0-8032 2-589 37 233-0 111-7 5. FouselOil,orAmy-) lie, . . .> CsHioO 0-8184 3-147 44 269-6 132-0 6. Hexylic or Caproic, C6H14O 0-833 3-53 51 299-309 138-154 7. Heptylic, C7H16O 0-819 58 851-0 177-2 8. Octylic or CapryHc, CgHisO 0-823 4-50 65 356-0 lSO-0 12. Laurylic, C.H,eO ... ... ... 1 16. Ethal 6r Cetylic, . CieH^O ... 27. Cerotiu or Cerylic, C27H560 30. Melissin or Mel- ) lisylic, . . \ C^oHcoO ... 1 ... ... Of these ethylic alcohol, wood spirit, and fousel oil are the three of most importance to the analyst. Ethylic Alcohol, CoHgO, specific gravity, 0-815 at 0°, 0-79381 at 15-5° ; boiling point, 78°-3. Absolute alcohol does not dissolve common salt, nor does it give a blue colour when digested with anhydrous sulphate of copper, if perfectly water-free. Filter jiaper saturated in the following solution — viz., two parts of § 228.] ALCOHOL. 467 citric and one of molybdic acids heated to incipient fusion, and dissolved in 30 to 40 pints of water and dried at 100°, is not bleached when soaked in absolute alcohol ; but should water be present, the blue of the paper is entirely discharged. Pure absolute alcohol burns with a white flame, but if water is present with a blue. There is no cloudy appearance when mixed with water, show- ing the absence of oily matters. It should be also perfectly neutral to test paper, and leave no residue on evaporation. It must be remembered that, in a commercial sense, " absolute alcohol " is any stronger spirit than can be obtained by ordinary distillation ; and, since this is the case, it would be most unwise for any action to be taken under the " Sale of Food and Drugs Act," unless a distinctly fraudulent statement has been made. " Absolute alcohol," as bought over the coimters of the chemist, is seldom above from 93 to 95 per cent. § 228. Rectified Spirit, as defined by our own pharmacopoeia, should be of specific gravity 0-83.8; by that of the Netherlands 0-830 to 0-834; of GermanV, Switzerland, and Norway, 0-8336; of Austria, 0-838; of France, 0-835 to 0-841. It should be neiitral, colourless, volatilising without residue, and free from other alcohols. Proof Spirit, — a term in constant use for purposes of ex- cise, — is a diluted spirit, which was defined by Act of Parlia- ment (58 George III.) to be "such as shall, at the tem])erature of 10° -0 [51° Fahr.] weigh exactly twelve-thii-teeuth parts of an equal measure of distilled water." According to Drink- water it consists of — Alcohol by weight, .... 49-24: Water by weight, . . . . 50-7G ioa-00 and its specific gravity at 15°-5 is 0-91984. In the analysis of all spirits (seeing that the terms "proof" and " under proof" are used and known in the trade), the state- ments of results should always include the percentage of proof spirit. Spirits weaker than proof are described as U.P., under -[jmof ; stronger than proof as O.P., over proof; thus a spirit of 50 U.P. means 50 water and 50 proof spirit, GO U.P., 60 water and 40 proof spirit. On the other hand, 50 O. P. means that the alcohol is of such a strength, that, to every 100 of the spirit, 50 of water would have to be added to reduce it to proof strength. In all the above the strengths are only good for the normal temperature of 15°-5. 468 FOODS : their composition and analysis. TABLE XXXIII.— ESTIMATION OF ALCOHOL, Specific gravity, lob". Absolute Alcohol by weight. 1 Per cent. Absolute Alcohol Dy \olume. Per cent. Proof Spirit. Per cent. Specific gravity, 15-6°. Absolute Alcohol ay weight. Per cent. Absolute Alcohol )v volume. Per cent. 1 Proof Spirit. Per cent. 1.0000 0-00 0-00 0-00 •9549 3r69 38-11 66-80 •9999 05 0-07 0-12 •9539 32-31 33-82 68-04 •99S9 0-58 0-73 1-28 -9529 32-94 39-54 69-29 •9979 112 1 -42 2-48 •9519 33-53 40-20 70-46 •9969 1-15 2-20 3^85 •9509 34-10 40-84 71-58 •9959 2 33 2-93 5-13 •9499 34-57 41-37 72-50 •9949 2^89 3-62 6-34 •9489 35-05 41-90 73-43 •9939 3^47 4-34 7-61 •9479 35-55 42-45 74-39 •9929 4-06 5-08 8-90 •9469 36-06 43-01 75-37 •9919 4-69 5-86 10-26 •9459 36-61 43-63 76-45 •99U9 5-31 6-63 11-62 •9449 37-17 44-24 77-53 •9899 5^94 7-40 12-97 •94:39 .37-72 44-86 78-61 •9889 6^64 8-27 14-50 •9429 38-28 45-47 79-68 ■9879 7-33 9-13 15-99 •9419 38-83 46-08 80-75 •9S69 8-00 9-95 17-43 •9409 .39-35 46-64 81-74 •9859 8-71 10-82 18-96 •9399 39-85 47-18 82-69 •9849 9-43 11-70 20-50 •9389 40-35 47-72 83-64 •9839 1015 12-58 22-06 •9379 40-85 48-26 84-58 •98-29 10-92 13-52 23-70 •9369 41-35 48-80 85-53 •9819 11-69 14-46 25-34 •9359 41-85 49-34 86-47 •9809 12-46 15-40 26-99 •9349 42-33 49-86 87 -.37 •9799 13-23 16-33 28-62 -9339 42-81 50-37 88-26 •9789 14 00 17-26 .-30-26 9329 43-29 50-87 89-15 •9779 14-91 18-36 32-19 •9319 43-76 51-38 90-03 •97C9 1575 19-39 33-96 •9309 44-23 51-87 90-89 •9759 16-54 20-33 35-63 •9299 44-68 52-34 91-73 •9749 17-33 21-29 37-30 •9289 45-14 52-82 92-56 •9739 18^15 22-27 39-03 •9279 45-59 53-29 93 -.39 •9729 18-92 23-19 40-64 •9269 46-05 53-77 94-22 •9719 19-75 24-18 42-38 •9259 46-50 54-24 95-05 •9709 20-58 25-17 44-12 •9249 46-96 54-71 95-88 •9099 21-38 26-13 45-79 -9239 47-41 55-18 96-70 •9689 22^15 27-04 47-39 •9229 47-86 55-65 97-52 •9679 22-92 27-95 48-98 •9219 48-32 56-11 98-34 •9669 23-69 28-86 50-57 •9209 48-77 50-58 99-16 •9659 24-46 29-76 52-16 •9199 49-20 57-02 99-93 •9649 •96.S9 25-21 25-93 30-65 31-48 53-71 55-18 •9198 49-24 57-06 100 -OOPS •9629 26-60 32-27 56-55 •9189 49-68 57-49 100-76 •9619 27-29 33-06 57-94 •9179 50-13 57-97 101-59 •9609 28 00 33-89 59-40 -9169 50-57 58-41 102-35 ■9599 28-02 34-61 60-66 •9159 51-00 58-85 103-12 •9589 29-27 35.35 61-95 •9149 51-42 59 ■26 103-85 •9579 29-93 36-12 63-30 •91.39 51-83 59-68 104-58 •9569 30-50 36-76 64-43 •91'29 52-27 60 12 105-35 •9559 3106 37-41 65-55 •9119 52-73 60-56 106-15 ESTIMATION OF ALCOHOL, TABLE XXXIII.— Continued. 469 Specific gravity, lo-S". Absolute Alcohol by weight. Per cent. Absolute Alcohol by volume. Per cent. Proof Spirit. Per cent. Specific gravity, 155°. Absolute Alcohol by weight Per cent. Absolute Alcohol by volume. Per cent. Proof Spirit. Per cent. •9109 53-17 61 ^02 106-93 •8649 73^00 79-54 139-39 9099 53-61 61-45 107-69 -8639 73^42 79-90 140-02 9089 54 05 61-88 108-45 -8629 73-83 80-26 140-65 9079 54-52 62-36 109-28 -8619 74-27 80-64 141-33 9069 55-00 62-84 110-12 •8609 74-73 81-04 142-03 9059 55-45 63-28 110-92 •8599 75-18 81-44 142-73 9049 55-91 63-73 111-71 •8589 75-64 81-84 143-42 9039 56-36 64-18 112-49 -8579 76-08 82-23 144-10 9029 56-82 64-63 113-26 •8569 76-50 82-58 144-72 9019 57-25 65-05 113-99 •8559 76-92 82-93 145-34 9009 57-67 65-45 114-69 •8549 77-33 83-28 145-96 8999 58-09 65-85 115-41 •8539 77-75 83-64 146-57 8989 58-55 66-29 116-18 •8529 78-16 83-98 147-17 8979 59-00 66-74 116-96 •8519 78-56 84-31 147-75 8969 59-43 67-15 117-68 •8509 78-96 84-64 148-32 8959 59-87 67-57 118-41 •8499 79-36 84-97 148-90 8949 60-29 67-97 119-12 -8489 79-76 85-29 149-44 8939 60-71 68-36 119-80 -8479 80-17 85-63 150-06 8929 61-13 68-76 120-49 -8469 80-58 85-97 150-67 8919 61-54 69-15 121-18 -8459 81-00 86-32 151-27 8909 61-96 69-54 121-86 •8449 81-40 86-64 151-83 8899 62-41 69-96 122-61 •8439 81-80 86-96 152-40 8SS9 62-86 70-40 123-36 •8429 82-19 87 '27 152-95 •8879 63-30 70-81 1-24-09 •8419 82-58 87-58 153-48 8869 63-74 71-22 124-80 •8409 82-96 87 -8 S 154-01 8859 64-17 71-62 125-51 •8399 83-35 88-19 154-54 8849 64-61 72-02 126-22 •8389 83-73 88-49 155-07 8839 65 04 72-42 126-92 -8379 84-12 88-79 155-61 8829 65-46 72-80 127-59 •8369 84-52 89-11 156-16 8819 65-88 73-19 128-25 •8359 84-92 89-42 156-71 8809 66-30 73-57 128-94 •8349 85-31 89-72 157-24 8799 66-74 73-97 129-64 •8339 85-69 90-02 157-76 8789 67-17 74-37 130-33 •8329 86-08 9U-32 158^28 8779 67-58 74-74 130-98 -8319 86 46 90 61 158^79 8769 68-00 75-12 131-64 -8309 86-85 90 90 159-31 8759 68-42 75-49 132-30 -8299 87 23 91 20 ir)9-82 8749 68-83 75-87 132-95 •8289 87-62 91 49 160-33 8739 69-25 76-24 133-60 •8279 88 00 91 78 160-84 8729 69-67 76-61 134-25 -8'269 88-40 92 08 161-37 8719 70-08 76-98 134-90 -8259 88-80 92 39 161-91 8709 70-48 77-32 135-51 -8-249 89 19 92 68 162-43 8699 70-88 77-67 136-13 -8239 89-58 92 97 162-93 8689 71-29 78-04 136-76 -8229 89-96 93 26 163-43 8679 71-71 78-40 ]37-40 -8219 90-32 93 52 163-88 8669 72-13 78-77 138 05 -8209 90-68 93 77 164-33 •8659 72-57 79-16 138-72 •8199 9104 94-03 164-78 470 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 229. TABLE XXXlIL—ConUnued. Specific Absolute Alcohol Absolute Alcohol Proof Specific Absolute Alcohol Absolute Alcohol Proof Spirit. Per cent. ->5Sf by weij>ht. Per ceni. by volume. Per cent. Spirit. Per cent. gravity, l5-5°. by weight. Per cent. by volume. Per cent. •8189 91-39 94-28 165-23 •8039 96-73 97-96 171-68 •8179 91-75 94-53 165 67 -8029 97-07 98-18 172-05 •81G9 92-11 94-79 166-12 •8019 97-40 98-39 172-43 •8159 92-48 95 06 166-58 •8009 97-73 98-61 172-80 •8149 92-85 95-32 167-04 -7999 98 06 98-82 173-17 ■8139 93-22 95-58 167-50 •7989 98-37 99 00 173-50 •8129 93-59 95-84 167-96 •7979 98-69 99-18 17384 •8119 93-96 96-11 168-24 •7969 99 00 99-37 174-17 •8109 94-31 96-34 168-84 -7959 99-32 99-57 174-52 •8099 94-66 9G-57 169-24 -7949 99-65 99-77 174-87 •8089 95 00 96-80 169-65 •7939 99-97 99-98 175-22 •8079 •8069 95-36 95-71 97-05 97-29 17007 170-50 Absolute Alcohol. •8059 96-07 97-53 170-99 •8049 96^40 97-75 171-30 •7938 100 00 100-00 175-25 § 229. Tests for Alcohol. — The principal tests for alcohol are the following : — (1.) Production of Acetic Ether. — To a distillate or aqueous solution sujiposed to contain alcohol, some acetate of soda is added and sulphuric acid in amount more than sufficient to decompose the acetate. The flask containing the mixture is connected -with a Liebig's condenser, placed vertically, and boiled for a few minutes; any volatile vapour is condensed, and f?ills back again into the flask. On removing the cork, if acetic ether lias been produced, it can readily be detected by its odour. (2.) Reduction of Chrortiic Acid or Bichromate of Potash to Oxide of Chro7nium. — A crystal of chromic acid, placed in a test-tube, with a fluid containing alcohol warmed to a boiling temperature, is decomposed into the green oxide of chromium. Instead of chromic acid, a test-solution of one part of dichromate of potash dissolved in 300 parts of sulphuric acid may be used. A portion of the liquid to be tried is mixed with twice its volume <)f concentrated sulphuric acid. On pouring a small quantity of this mixture into a quantity of the test-solution, a deep green )s prodviced where one fluid touches the other. This is a xerj good test in the absence of other reducing agents, such as formic acid, ether, ifec. (3.) Dr. Edmund Davy's Test. — Dr. Davy has i)roposed a test for alcohol founded on a colour reaction, and produced also by PURE BRAZILWOOD. '■iTl POBE FDOHSINE. '■\T] BLACK ELDEB, pure. DWARF ELDEIl.t P«rll»lly iy greenish -grey ■.ra"'^n ff-' ""' NOTES TO TABT.E A. 1. Each wine re-acts in a slightly different mamier, according to its variety, «ge, Ac Tliia Table refers to wines iif five to fifteiii raontlia old, and particularly ti. tlio following ;— Pinot, Carignane, Teinturier, Carlieiiet. 2. The word pure means not mixed with wine. The reactions shown were obtained by acting upon solutions of the substances in water, containing 10 per cent, of alcohol, and were made of such strength that the colours corresponded in iutensity with those of the wines being examined. 3. Brazil-wood, 1 part, wine i parts, means that tBe intensity of the colour of the liquid examined, resulted from the mixture of the decoction of Brazil-wood and of wine in the proportions named. These proportions refer only to the intensity of tin colotation, and represent but a very minute ponderable quantity of the adulter- ating substance. § 2 2D.] TESTS FOR ALCOHOL. 471 methyl, propyl, butyl, and amyl alcohols, ether and aldehyde. A solution of one part of molybdic acid in ten of strong sulphuric acid, is warmed in a porcelain capsule, and the liquid to be tested allowed to fall gently on it. If alcohol is present, a, blue coloration appears either immediately, or in a few moments; the liquid gradually absorbs moisture from the air, and the colour disappears, but it may be reproduced by evaporation.* (4.) The production of Iodoform. — According to Lieben, one part of alcohol in 2000 of water can be detected by adding to some of the warmed liquid a few drops of a 10 per cent, solution of soda, and dropping in a solution of potassiiim iodide, fully saturated with free iodine, until the liquid is yellowish-brown ; then the alkali solution must be added until the whole is colourless, and the mixture allowed to stand for many hours, when a yellowish crystalline deposit of iodoform is obtained. Under the micro- scope the latter presents the appearance of hexagonal plates, six-rayed, or other varieties of, stellar crystals : CoHeO H- 4I2 -)- 6NaH0 = CHI3 + NaCHOo -f 5NaI -f SHoO. The objection to this test is, however, that other alcohols, aldehyde, giun, turpentine, sugar, &c., give a- similar reaction, f (5.) Tlte production first of Acetic Acid, then of Kakodyl. — A very delicate test for alcohol, and one specially suited for its detection in the blood, &c., is recommended by Bucheim. The finely-divided substances are put in a tubulated retort, and, if acid, carefully neutralised. In the neck of the retort is placed a little porcelain or platinum boat, containing platinum black, and at each end there is a moistened piece of strongly- blued litmus-paper. On warming the retort, if alcohol be present, it is oxidised by the platinum black to aldehyde and acetic acid; hence, the hinder piece of litmus-paper will be reddened, the front one unchanged. If only a drop of acetic acid be present, it is possible to detect it in the following way: — The i)latinum black is washed, the washing water neutralised with potash, and dried after the addition of a few grains ei arsenious acid. On warming the dry residue in a small glass tube, if even a very small admixture of acetic acid be present, the smell of kakodyl will be perceptible. (6.) The Action of Alcohol on Benzoyl Chloride. — This test, pro- posed by Berthelot, is based on the fact that veiy small quantitie.s of alcohol decompose benzoyl chloride with the formation of ethyl * Proceedings of the Bo;/al Irish Academij [2] ii. 579-5S2; Journ. Cliem. 80c., 7, 1S77, p. lbs. Gladstone and Tribe (Journ. Chevi. Soc, No. ccxlix.) have shown that the test is unreliable, for the same reaction takes place with ammonium sulphide, sodium sulphite, formic acid, sugar, and, in fact, with most reducing agents. t Rajewsky has found that the brain of a rabbit, which had been starved for two days before death, gave a marked reaction for alcohol with the iodoform test, and the same result was obtained from the muscles and tissues of rabbits. He therefore considers that alcohol always exists in the animal organism, or that it is produced during distillation. — PJliiger's Archiv.filr Ph)j.iiologie, xi., 122, 127. 472 FOODS : their composition and analysis. [§ 230. benzoate. Tlie sample is shaken up in the cold with a few drops of benzoyl chloride; any ethyl benzoate formed sinks to the bottom with the excess of benzoyl chloride. This heavy layer is removed by a pipette, and heated with a little caustic soda or potash, which dissolves at once the benzoyl chloi-ide, but not the ethyl benzoate, and the latter may be recognised by this insolubility, by its general properties, and by its boiling point. It is nearly insoluble in water, burns with a smoky flame, has a characteristic pleasant odour, and boils at 213°. Mr. J. Hardy has proposed a very simple test for the detec- tion of alcohol, which may be performed as follows : — Two common " Nesslerising " glass cylinders are taken, and a little guaiacum resin, which has been removed from the interior of a freshly broken lump, is shaken up with the sample to be tested. The' liquid is filtered, and a few drops of hydrocyanic acid and a drop of very weak solution of sulphate of copper added. Exactly the same process is adopted with an equal bulk of distilled water. On now placing the two liquids in the glass cylinders side by side, over a porcelain plate, the liquid to be tested, if it con- tains alcohol, Avill be found of a blue colour decidedly darker than that of the distilled water. § 230. Sejjaration of Alcohol from Animal Matters. — In order to obtain alcohol from organic matters {e.g., the contents of the stomach, or the tissues), the following process will be found con- venient : — Solid matters, such as the tissues, are cut up as finely as possible, and placed with water in a retort attached to a sviit- able condenser. Most liqiiids require no previous preparation, and are merely poured into the retort or flask, as before de- scribed ; but it is desirable in the treatment of urine to add a little tannic acid. About one-third to one-half of the liquid is distilled over into a flask closed by a mercury valve. The pro- duct is now exactly neutralised with decinormal alkali to fix any volatile acid, and again slowly distilled, about one-third being drawn over. The liquid is next neutralised with sul- phuric acid, to fix volatile alkalies, and redistilled. This final distillate contains all the alcohol, but neither volatile acids nor alkalies. The liquid thus obtained may even now be too dilute to respond conveniently to tests, and it may therefore be digested for some hours with a little caustic lime, and then very slowly distilled. The distillate should finally be carefully measured or weighed, and divided into two parts, one of which serves for the application of the usual tests, the other (if alcohol be found) can be oxidised in the manner described at p. 478, and estimated as acetic acid volumetrically. J. Bechamp {Compt. Rend., 89, 573, 574) has succeeded in § 231.] ESTIMATION OF ALCOHOL. 473 obtaining a sufficient quantity of alcohol from the fresh brain of an ox to estimate by specific gravity, and has also separated ife from putrefying animal matter. In fact, it is capable of proof, that all putrefaction is accompanied by the production of minute quantities of alcohol, and that the living cells of the body also* produce it. Hence, in questions of poisoning, it is not enough to obtain qualitative reactions for alcohol, but the quantity also- must be accurately estimated. ESTIMATION OF ALCOHOL AND ANALYSIS OF ALCOHOLIC LIQUIDS. § 231. Excellent tables for the use of analysts have beeiE published both by Mr. Hehner and by Dr. Stevenson. The table on p. 468 et seq. will be found, in the absence of the tables mentioned, sufficient for ordinary use ; any specific gravity not given can be intercalated by the ordinary rules of arithmetic. Another method, sometimes called Groning's of arriving at the strength of dilute spirits, is based on the fact that the temperature of the vapour is an exact measure of the strength of the alcohol. The bulb of a thermometer is put (on the small scale) into a flask with a bilateral tube, and the temperature of the vapour carefully noted. The following table (XXXIV.) may be used. The boiling-point is also a useful guide ; for within certain, limits the boiling-point of alcoholic liquids is not materially altered by admixture with saline and organic matter. A ther- mometer with a movable scale is employed. Before using it, the thermometer is immersed in boiling distilled water, and the 100° [212° Fahr.] of the scale accurately adjusted to the level of the mercury ; it is then ready for an operation of several hours, or even an entire day, if no considerable varia- tions of atmospheric pressure are experienced. The methods used in the Municipal Laboratory, Paris, for the analysis of alcohols are the most practical and the easiest to apply to the small quantities of alcohols submitted to an analyst under the Sale of Food and Drugs Act, and have been made considerable use of in the following pages. The estimations of the higher alcohols, of aldehydes, of fur- furol, and so forth, are to a great extent determined by colour 474 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 231. reactions, the alcohol to be tested being always brought to the definite strength of 50°; by this device variations in colour, owing to vai'ying strengths of alcohol, are avoided. The French chemists estimate the following: — (1.) Alcohol; (2.) acidity, expressing it as acetic acid; (3.) aldehydes; (4.) ethers, returning the ethers as ethyl acetate; (5.) higher alcohols, returning them as isobutylic alcohol ; and (6.) bases. The scope and character of such an analysis may be gathered from the following example of the analysis of a sample of brandv : — Brandy made from Wine. Sp. gr. at 15°, 0-9414 Alcohol, per cent, (volume), . . . . 4S"2 Extract, per litre, . . . . . . I2'G4 grms. r Saccharose, per litre, , . . . 8 '2 ,, Sugars - llnverted sugar, per litre, . . . 3 '7 ,, Colour, Tannin and caramel. Grammes. Proportion of each Impurity, per cent, of Total Impurity. Per Litre of Brandy. Per cent, of Absolute Alcohol. Acidity, as acetic acid, . Aldehydes, as acetic aldehyde, Furfurol, ..... Ethers, expressed as ethyl acetate. Higher alcohols, expressed as iso- ' butylic alcohol, . J Ammonia, amides, \ Alkaloids, pyridine bases, &c.. 0-3360 0-0620 0-0027 0-2024 0-3990 0-0350 0-0056 0-0697 0-0130 0-0005 0-0419 0-0S28 0072 0011 32-23 6-01 0-26 19-38 38-29 3-33 0-50 Coefficient of impurities, per cent. \ of absolute alcohol, , . f 0-2162 100-00 231.] ESTIMATION OF ALCOHOL IN SPIRITS, ETC. 475 TABLE XXXIV., showing the Alcoholic Content by Volume OF Boiling Spirits and of their Vapour, prom the Temperature of the latter, as observed by a Ther- mometer. By Groning. Temperature of the vapour Alcoholic Content of the Distillate. Per cent. Alcoholic Content of the Boiling Liquid. Per cent. Temperature of the Vapour Alcoholic Content of the Distillate. Per cent. Alcoholic Content of the Boiling Liquid. Per cent. 170-0 171-8 172-0 172-8 174 174-6 176-0 178-3 180-8 183-0 185 187-4 93 92 91 904 90 89 87 85 82 80 78 76 92 90 85 80 75 70 65 50 40 35 30 25 189-8 192 194-0 196-4 198-6 201-0 203-0 205-4 207-7 2100 212-0 71 68 66 61 55 50 42 36 28 13 20 18 15 12 10 7 5 3 2 I TABLE XXXV., exhibiting the Boiling Points of Alcohol and Water of the Given Strengths. By Groning. Boiling-point (F.) Alcohol per cent. Boiling-point Alcohol per cent. per Volume. per Volume. 205-34 5 179-96 55 199-22 10 179-42 60 195-80 15 178-70 65 192-38 20 177-62 70 189-50 25 176-54 75 187-16 30 175-46 80 185-00 35 174-92 85 183-38 40 174-20 90 182-12 45 173-14 95 181-58 50 172-00 100 476 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 232. § 232. In the examination of alcoholic liquors, one of the analyst's first steps is to determine the percentage of alcohol, and the methods by which this is done are equally applicable (with slight modifications) to all liquids containing alcohol. The percentage is ascertained — (1.) By distillation, and taking the specific gravity of the- distillate. (2.) By Tabarie's method, applied especially to wines and beers. (3.) By Geissler's vaporimeter. (4.) By oxidation into acetic acid, and by several other methodSy which are, however, not much in use by the analyst. (1.) Distillation. — A convenient quantity {e.g., 100 cc. of beer or wine, 50 cc. of spirits, measured at 15°-5) is placed in a flask (a, fig. 62), having an angle tube connected by means of a cork to a Fig. 62. Liebig's "condenser. The distillate is received in a flask {h) pro- vided with a doubly perforated cork, into which the bent tube of the condenser, as well as a tube provided with a mercury valve, to prevent loss, is adjusted ; the latter may be readily made bv putting a very small quantity of mercury into the bend of an ordinary thistle-head funnel (c). This precaution is only neces- sary when very small quantities are operated upon. Experiments with 50 cc. of spirit distilled into a flask unprovided with a valve, have shown that there is no appreciable loss; but distillation ESTIMATION OF ALCOHOL IN SPIRITS, ETC. 477 § 232.] into an open vessel will always give results far too low. Beer and wine yield the whole of their alcohol when half is drawn over ; spirits should be distilled nearly to dryness. In any case, the distillate should be made up to exactly the same bulk as the original liquid at the same temperature, its specific gravity taken in a proper specific gravity bottle, and the percentage of alcohol obtained by reference to the tables given at p. 468 et seq. Spii-its are best returned as containing so much proof sjnrlt, by weight and by volume ; wines and beers, so much alcohol per cent., by weight and by volume. (2.) Taharies method, when properly performed, is suflaciently accurate for all practical purposes in the case of beers, wines, and similar liquids. The specific gravity is first accurately taken at 15°-5 ; a measured quantity— say 100 cc— is then boiled long enough to evaporate away the whole of the alcohol, made up to the original bulk at the same temperature, and its specific gravity again determined. From these data the specific gravity of the liquid, which, if it had been condensed, would have collected in the flask before-mentioned, is determined. Thus, specific gravity of the liquid before boiling, divided by the specific gravity of the de-alcoholised liquid = specific gravity of the diluted alcohol which has been boiled away. An actual example will sufiice : — A beer, before boiling, had a specific gravity of 1-014 ; after boiling, and on making it up to the original bulk, its specific gravity was 1-0172 ; now J-oVts" = '9968, and on refer- ence to the table at p. 468, -9968 is found to cori-espond to 1-7 per cent, of alcohol. (3.) Geissler's Vaporimeter is capable of giving sufliciently accurate results for tech- nical purposes, and as it has the advantage of great expedition, it may always be used to supplement and check other methods which take more time. It depends on the measurement of the tension or elastic force of the vapour of the liquid, as indi- cated by the height to which it raises a column of mercury. The apparatus (see fig. 63) consists of four parts- — viz., (1.) A brass vessel A, containing water; (2.) a doubly bent tube, BB, fastened to a scale ; (3. ) a cylindrical vessel C, which is filled with mercury and the fluid to be tested ; (4.) a brass vessel D, in the uppper part of which there is a Ficr. C3. 478 FOODS : their composition and analysis. [§ 232, thermometer. By removing the bent tube and its connections from A, it may be turned upside down, and C detached ; the alcoholic fluid is then poured in so as to fill the space between c and d, which, when the instrument is inverted, is empty. Lt is now- connected with the bent tube, and adjusted exactly as in the figure, the water made to boil in A, when the mercury runs up to a certain height in the tube, and the percentage is directly read from the scale. Shoidd the thermometer not register 100°, certain corrections must be made, which is most conveniently done by a table, sold with the instrument. Care in this manipu- lation must always be taken to exclude air from the bulb. The author has found it always necessary to test various points of the scale with known mixtures of alcohol and water, and to draw up a table of corrections. "With this precaution, it will be found a most useful instrument. (4.) Oxidation into Acetic Acid is an extremely accurate method of determining vinic alcohol, and is specially applicable to small quantities. Dr. Dupre recommends the following process* : — A small quantity of the distillate, representing about a gramme of alcohol, is j)ut in a small strong assay-flask, and mixed with 10 cc. of an oxidising solution, composed of 147 grms. of dichro- mate of potash and 220 grms. of sulphuric acid, made up to 1,400 cc. by water. The flask is well stoppered by caoutchouc, and firmly tied down by canvas and string. It is then suspended upright in a water-bath (the neck being above the water), and heated for two hours between 80° and 90°. The flask is next removed, and the excess of dichromate reduced by zinc and sul- phuric acid ; the solution is transferred to a small retort (adding some sulphuric acid and bits of tobacco-pipe), and distilled over from a spermatceti-bath (see fig. G2). It will be found necessary to distil at least thrice nearly to dryness, each time adding water to the contents. The united distillates contain acetic acid, the result of the oxidation of the alcohol. This acetic acid may be determined by a volumetric solution of soda, and the amount of alcohol to which it is equivalent calculated by the following short table : — Acetic Acid. Alcohol. 1 .... = ... . -7666 2 . . , . = .... 1-5332 3 .... = ... . 2-2998 * It has been shown by Wanklyn that alcohol may also be oxidised into acetic acid very readily by an alkaline solution of permanganate of potash ; it would appear that in this case thei-e is no previous formation of aldehyde. A. Letteltur also finds that an ammoniacal solution of copper-oxide at ISC'* has the same effect. Compt. Tiend., 89, 1105. Alcohol. Acetic Acid. Alcohol. 3-0664 8 = 6-1328 3-8330 9 = 6-8994 4-5996 10 . = 7-6666 5-3662 § 233.] ESTIMATION OF ALCOHOL IN SPIRITS, ETC. 479 Acetic Acid. 4 6 '. = 7 There are several other metliods of estimating alcoliol, but the above are the most practical and efficient ; and whenever the amount of alcohol is important (as, for example, in the case of spirits) the analyst should determine it in at least two different ways. The specific-gravity methods presuppose the presence of ethylic alcohol only ; but it is often necessary to test fluids alsa for methylic and amylic alcohols. For these the following pro- cesses are available : — § 233. Methylic Alcohol. — 100 cc. of the suspected spirit are distilled twice, having been rendered alkaline during the first process and acid during the second, about two-thirds being dis- tilled over each time. The distillate is now shaken up with dry potassium carbonate, and, after standing over night, the upper layer is taken oft' by a syphon or pipette, and again twice distilled, about 15 cc. being driven over. This will contain any methylic alcohol present in the original 100 cc. A portion of the distillate is^ now diluted with water to a strength from 10 to 15 per cent., and in this diluted spirit the alcohol determined — (1.) By specific gravity ; (2.) by Geissler's vaporimeter ; and, (3.) by oxidation into acetic acid. If ethylic alcohol alone is present, all three methods fairly agree. The specific gravity will give the total amount of both alcohols, the specific gravity of aqueous methylic and ethylic alcohols being almost identical ; but since methylic alcohol has a higher vapour tension than ethylic, Geissler's vaporimeter will give a higher result. The oxidation jjrocess will, on the other hand, give a lower result, for methylic alcohol yields water and COg, so that the acetic acid found is derived wholly from the ethylic alcohol, and the diff"erence between the strength thus found and that derived from the specific gravity gives a rough indication of the proportion of methylic alcohol present. If the methylic alcohol is in sufficient quantity, instead of the usual slight vacuum on opening the flask, there is an escape of carbonic anhydride, and there is no reason why this gas- should not be either absorbed or collected and estimated. Dr. Dupre gives the following example. A pure whisky- showed — Strength by specific gravity, ... 9 -S3 per cent. ,, Vaporimeter, . . . 9-/5 ,, ,, Oxidation, , . . . 9-75 ,, * Hehner has raised a doubt as to the usefulness of the vaporimeter for the quantitative estimation of methylic alcohol ; in the oxidation process he prefers to estimate the quantity of chromate reduced rather than the acetic &cidi.— Analyst, Feb. 7, 1SS7. 480 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 233. The same whisky, adulterated with 10 per cent, of ordinary methylated spirit, and tested, gave — Strength by specific gravity, . . 10 "OS per cent. ,. Vaporimeter, . . . 10'45 ,, Oxidation, . . . 9-50* ,, The remainder of the distillate may be used in producing methyl- aniline violet or oxalate of methyl. The general process for the production of methyl-aniline violet is as follows : — 10 cc. of alcohol, with 15 grms. of iodine and 2 grms. •of red phosphorus, are put into a small flask, and distilled into 30 or 40 cc. of water. The alcoholic iodide which settles to the bottom is separated by a pipette, and collected in a flask contain- ing 5 cc. of aniline. If the action be too violent, the flask can. be cooled with cold water ; if too slow, a little heating may be necessary. At the end of an hour the crystals are dissolved in hot water and boiled, an alkaline solution is afterwards added, and the bases rise to the top in the form of an oily stratum, which may be separated by bi'inging the oil, by the addition of water, on a level with the neck of the flask. The oxidation of the bases may be eff'ected in various ways, but best by pouring a ■cubic centimetre of the oily liquid on 10 grms. of a mixture formed of 100 grms. of quartz sand, 2 of chloride of sodium, and 3 of nitrate of copper. After incorporation it is introduced into a glass tube, and kept at 90° in a water-bath for eight or ten hours; it is ultimately exhausted by warm alcohol, thrown on a filter, and made up to 100°. If the alcohol was pure the tint is red; if it contained 1 per cent, of methyl alcohol the colour, by the side of the preceding, is manifestly violet ; with 2-5 per cent, the shade is a very distinct violet ; and with 5 per cent, it is considerably dai-ker. The pi-ocess may be made quantitative by having volumetric solutions of methyl and ordinary alcohol. Very minute quantities of methyl alcohol may be detected by adding 5 cc. of the liquid to 95 cc. of water, and then again •diluting 5 cc. of this liquid with 400 cc. of water, and heating it in a porcelain capsule. Fragments of white merino (free from sulphur), immersed in the liquid for half an hour, will remain white, if the alcohol was pure ; if methyl was present, they will be of a violet tint.f Oxalate of Methyl may be obtained by mixing the distillate with half its bulk of sulphuric acid and double the quantity of hydropotassic oxalate. The whole should stand in the retort for * Note on the Examination of Whisky and other Spirits for Methylated •Sijirit and Fousel Oil, by Dr. Dupre'. Analyst, i., 1876, p. 4. + A. Riche and Brady, Comptes Rendus, vol. xxx., p. 1096. § 233.1 ESTIMATION OF ALCOHOL IN SPIRITS, ETC. 481 twenty-four hours, and then be distilled. Crystals appear after a time in the cooler parts of the flask or retort ; their composition is (CH3)oCo04. It would also be quite possible to produce such compounds as the salicylate of methyl, &c.* Formic Acid. — It has been already stated that methyl alcohol, when fully oxidised, is resolved into COg and water. This takes place in two stages ; first, formic acid is produced, and then this formic acid breaks up. Thus: — (1.) Methyl Alcohol, Formic Acid. CH4O + O2 = CH2O2 + H2O (2.) Formic Acid, 2CH2O2 + O2 = 2CO2 + 2H2O It hence follows that if a spirit be distilled in the manner recommended by Dr. Dupre, and only partially oxidised, it is possible to get formic acid, which has some very characteristic properties. To obtain this result a small portion of the dis- tillate, 2 to 4 cc, is taken, and 3 grms. of potassic dichromate are added, with an equal quantity of pure sulphuric acid, and four or five times as much water. This is allowed to act for twenty minutes, and then distilled ; the liquid is alkalised with sodium carbonate, evaporated to half its bulk, acidulated with acetic acid, transferred to a test-tube, and then heated gently with a 5 per cent, solution of nitrate of silver. If formic acid has been produced, there is a distinct precipitate of metallic silver. Acidity of Alcohols. — A definite quantity of alcohol is coloured by phenol-phthalein, and decinormal soda dropped in until the liquid is neutral ; the result may be expressed as acetic acid. Aldehydes. — Aldehydes are detected by a solution of fuchsine made as follows: — Bisulphite of soda solution (sp. gr. 1-3082), 100 cc. ; aqueous solution of fuchsine (1 per 1,000), 150 cc. ; pure sulphuric acid (66°), 15 cc. j the whole made up to a litre with distilled water. 4 cc. of this test are added to 10 cc. of a bitter principle, a crystalline substance — Angelicine — a resinous substance, an essential oil, and other constituents. Angelic Acid, CgHgO.,, forms transparent glittering prisms and needles, melting at 44" to 45*^ into an oil, which may be solidi- fied at 0° into a crystalline solid. If the heat be raised up to- 190° it boils and distils unchanged; it is inflammable, burning with a luminous flame. The acid reddens litmus, and has th© odour of the root. It scarcely dissolves in cold, but is soluble in liot water, in alcohol, ether, turpentine, and the fatty oils. It forms salts with bases, which lose a part of the acid on evapo- ration. It precipitates lead and silver salts white ; iron salts, dark yellow; and copper, bluish. By the aid of hydric iodide and red phosphorus, acting at 180" to 200°, angelica acid is changed into valerianic. Melting the acid with KHO decom- poses it into propionate and acetate of potash. Angelicine is, according to Brunnei', probably identical with hydrocarotin, a principle described by Husemann, found in the DoMcus carota, L., and to which the following formula is ascribed, C^gllgoO. Hydrocarotin, or angelicine, forms colourless, large, thin plates, without smell or taste, swimming in water, and becoming at 100° hard and brittle. At higher temperatures (120° -5) it melts without loss of weight to a yellow fluid, which solidifies as a resinous mass, and cannot be again crystallised. It is readily soluble in ether, chloroform, carbon bisulphide, benzine, oil of turpentine, and warm olive oil. It is not changed in colour by concentrated hydrochloric acid ; fuming nitric acid dissolves it with the evolution of gas. Concenti-ated sulphuric acid dissolves it to a red fluid, depositing brownish-white flakes on dilution with water. Angelica Oil is colourless, and lighter than water ; it has a penetrating odour and camphor-like taste, and resinifies on exposure to the air.* § 244. Oil of Coriander is a pale yellow oil, smelling like the fruit; of specific gravity -871 at 14°, and a portion distilling over at 150°. The volatile part corresponds to the formula. * Bibliography. Alscher. — Ber. d. deutsch. Chem.. Ges., 1869, 683. Brunner, Carl.— iV. Bep. Pharm., xxiv. 641-665. BncHNER, A. — Bepert. Pharm., Ixxvj. Chiozza. — Ann. Ckim. Phys. [.3], xxxix. 435. Husemann. — "Die Pflanzenstoffe." Berlin, 1871. Jaffe. — Ann. Chem,. Pharm., cxxxv. 291. Mayer and Zenner. — Ann. Chem. Pharm., Iv. 317. Eeinsch.— /a/trZ). Pharvi., vii. 79. § 245, 246.] GIN. 495 C^oHjgjHgO; the portion of a higher boiling point to 4C\oH^g,H^O. If both portions are distilled with phosphoric anhydride, a powerfully odorous camphor, CjoH^g, is produced.* § 24-5. Oil of Juniper is contained in the unripe beri'ies of the common juniper, in the proportion of from -4 to "75 per cent. It is colourless, or of a pale yellow, dissolving with turbidity in twelve parts of alcohol of S3 per cent. Miscible in all pro- portions with ether and bisulphide of carbon. Smell and taste mildly aromatic. Specific gravity 0-862 to 0-874, but the poorer commercial samples often have a specific gravity of 0-860. The perfectly colourless oil does not fulminate with iodine, but the commoner kinds explode powerfully.t If from 5 to 6 drops of th& oil be placed in a test-tube, and five times its bulk of sulphuric acid be added, much heat is developed with the evokition of vapour, and the iluid becomes dark yellow-red and turbid ; on now diluting wnth 10 cc. of 90 per cent, alcohol, the colour changes, to a somewhat dirty rose tint. The pure oil boils between 140° and 150°; it polarises to the left. On exposure to the air, oxygen is absorbed ; and on long standing, colourless tables of juniper camphor are separated. This camphor melts and sub- limes without decomposition, is easily soluble in ether and alcohol, and may be obtained in feathery crystals. The action of warm water on juniper oil, if kept up for some considei-able time, results in the formation of a crystalline hydrate. Oil of juniper is officinal in all the Continental phai'- macopceias, as well as our own. In such large doses as from 15 to 30 grms., it is fatal to kittens, apparently acting in the same way as turpentine. J § 246. Analysis of Gin. — The analyst should find in good gin at least 80 per cent, of proof spirit, and a variable amount of sugar and fiavouring- matters, seldom much over 5 or 6 per cent. Sul- phuric acid, sulphate of zinc, alum, and lead should always be looked for. Many writers seem to imagine that grains of para- dise is an adulterant. It is, however, in its properties merely a. * HtJSEMAXif. — "Die PflanzenstofFe." Berliu, 1871. Kawalier.— /oin-n. Pract. Chem., Iviii. 226. + It is said that the oil from the unripe fruit explodes, that from the fully ripe lierries losing this property. :J: Bibliof/raphy. HiRSCH.— "Priifung der Arzneimittel." Berlin, 1875. HusEMAXN.— "Die'Pflanzenstoife." Berlin, 1871. Simon.— "De Olei Juniperi Berol.," 1841. Berlin 31ed. Ver. Ztg. 19, 1844. Steer.— C7(cm. Centralhl, 1856-60. SouBEiRAX et Capitaine. — Journ. Pharm. [2], xxvj. 73. Zaubzer. — Repert. Pharm., xxii. 415. 496 FOODS : their composition and analysis. [§ 247. pepper, and much nonsense has been talked about it. It is •very doubtful whether any just conviction would be obtained for the addition of any harmless flavouring to the spirit ; nearly all pi'osecutions hitherto have been for dilution, and for dilution ■only. It appears that no genuine gin * is sold to the retailer 22 under proof, but the standard fixed by the 6th section of the Amended Act (see p. 49) lays down the limit of 35 degrees under proof, and anything below this must be returned by the analyst as adulterated. The alcohol should be determined by distillation, as before described (p. 476), and the percentage in the distillate estimated by specific gravity, and, if necessary, in other ways. Neither methyl-alcohol nor fousel-oil appears to have been found in gin. The residue after the distillation may be treated with petro- leum ether, benzine, &c., as in Dragendorfi^'s process for the test- ing of beers. The essential oils will be taken up by the petroleum ether, and may be identified by their odour and taste, and (if ■enough is obtained) by their physical properties. Sulphuric acid if in a free state, may be separated by cinchonine, as recommended under the article " Vinegar." The detection of alum, lead, and zinc is elsewhere described. ARRACK. § 247. The best qualities of arrack are manufactured by distillation of the fermented juice of the cocoa-nut tree, palmjrra tree, and other palms ; the coarser kinds are made from the distillation of fermented rice liquor. Arrack is nearly colourless, a slight tinge of yellow or brown being only observed in samples kept in casks for a length of time. The average strength and -composition of arrack are as follows : — * In an appeal case (heard before the legal limit was fixed) before Baron Cleasby and Mr. Justice Grove [Fashler v. Stevenilt), the analyst proved that the gin was 44 degrees below proof. The judges affirmed the conviction. Baron Cleasby thought the conviction was right. When the respondent asked for gin he meant such gin as is ordinarily sold ; and to sell him such gin as that in question was to sell, to the prejudice of the purchaser, gin which was not of the quality demanded. The amount of water proved to have been discovered with the gin afforded evidence that it had been added for the purjjose of fraudulently increasing its measure. Mr. Justice Grove concurred ; in his opinion, when it was proved that the gin contained so much more water than gin as ordinarily sold, the onus was thrown on the seller of proving that he was not aware of the state in which it was. § 248, 249. ABSINTHE. 497 Per cent. Specific gravity, . . . '9158 Alcohol, 52-700 Extract, -082 Ash, -024 Water, 4719t The Hindoos and Malays consume large quantities of arrack. It would appear in the East to be occasionally dragged, and Indian hemp and the juice from solanaceous plants are said to be employed for this purpose. LIQUEURS OR CORDIALS. § 248. The terra cordial, liqueur, &c., is applied to a number of liquids which essentially consist of very strong spirit, flavoured with essences, and often very brightly coloured by vegetable colouring-agents, such as turmeric, cochineal, &c. Occasionally injurious colours are used, and salts of copper, picric acid, and impure aniline dyes have been detected. Of the liqueurs, absinthe is the most important in relation to health, and will be considered separately. The alcoholic strength and general composition of a few others are as follows : — Specific gravity. Alcohol by weiglit. Extract Cane Sugar. Ottier Exirac- tive Matters. Ash. Bonekamp of Maa£;bitter, . •9426 42-5 2-05 0-106 Benedictme-bitter, . 1-0709 44-4 36-00 32-57 3-43 0-043 Ingwer, .... 1-0481 40-2 27-79 25-92 1-87 0141 Creme de meuthe, 1-0447 40-7 28-28 27-63 0-65 068 Anisette de Bordeaux, 1-0847 35-2 .34 82 34-44 0-38 0-040 Cura^oa, .... 1-0300 47-3 28-60 28-50 010 040 Mfimmel -liqueur, 1-0830 28-0 32 02 31-18 84 0-058 Pfeffer-munz liqueur, 1-1429 28-6 48 25 47-35 0-90 0-068 Swedish Punch, 1-1030 21-6 3G'61 ABSINTHE. § 249. Absinthe is a yellowish-green liqueur, which contains, as a peculiar and distinctive ingredient, a poisonous oil having a deleterious action on the nervous system. This, '-loorm- 33 498 foods: their, composition and analysis. [§ 249. wood oil,^^ is the produce of the ^'■Artemisia absinthium." Other flavouring oils are always present, such as peppermint, angelica, cloves, cinnamon, and aniseed. The green colour is produced by the juice of spinach, nettles, or parsley, or, in other words, it is due to chlorophyll. The absorption-spectrum of properly made absinthe, is the same as that of chlorophyll. Most samples of absinthe contain sugar. The average composition of tlie liqueur as consumed in London (where its use is on the increase) is as. follows : — Per cent. Absolute alcohol, 50-00 Oil of wormwood, -33 Other essential oils, ...... 2 'Si Sugar, I'oO Chlorophyll, ....... traces Water, 45 •Go On diluting absinthe the essential oils are thrown out of solution, and the liquid becomes turbid. The reaction is always slightly acid, due to a trace of acetic acid. Adulterations of Absinthe. — The composition of absinthe appears to be fixed by no definite standard of strength ; therefore, practically, the analyst has to look only for such substances as injurious colouring-matters and metallic impurities. Sulphate of indigo with turmeric is not unfrequently employed as a colouring agent, and similarly picric acid has been detected, and salts of copper. The latter is readily discovered by diluting the liqueur and adding ferrocyanide of potassium, which, if copper be present, will give a brown colouration; picric acid and indigo are detected in the way elsewhere described. (See Index.) Analysis of Absinthe. — The alcohol may be determined by dis- tillation, after diluting the liqueur to cause the oils to separate, and getting rid of some portion by filtration. To make an estimation of the essential oils, a measured quantity of the liqueur is diluted to twice its volume by the addition of water ; carbon disulphide is added, and the mixture shaken up in the tube described at p. 69. The carbon disulphide dissolves all the essential oils, and on evaporation leaves them in a state pure enough to admit of their being weighed. Absinthe, when taken habitually and for a lengthened time, produces a peculiar train of nervous symptoms which the French physicians afi^ect to dis- tinguish from the similar symptoms produced in inebriates by alcohol. In epilepsy caused by indulgence in absinthe, M. Voisin states, as the results of clinical observation, that the number of fits is far greater than in alcoholic epilepsy. FERMENTATION : FERMENTED LIQUORS. " The chemical act of fermentation is essentially a correlative phenomenon of a vital act beginning and ending with it. I think that there is never any alcoholic fermentation without there being, at the same time, organisation, development, and multiplication of globules, or the con- tinued consecutive life of globules already formed." — Pasteur. § 2o0. Fermented liquors are those mannfactured by fermen- tation — i.e., a peculiar, low vegetable growth has been allowed to grow and mxiltiply, assimilating material in one form, and excreting it in another, and this process has evolved, with other by-products, carbon dioxide and alcohol. Fermentation used to be considered as a sort of special action ; but modern research recognises a great variety of ferments, and it is a question whether all animal and vegetable cells, in the exercise of tlieir normal functions, do not act in a similar manner to the environing fluids, as do the organisms more distinctively called '• ferments." In normal alcoholic or spirituous fermentation, as in the manu- focture of beer, a minute unicellular plant, " yeast," grows and multiplies, and splits up the sugar, as in the following equation : — CcHi206=2C2H60 + 2C02, that is, one molecule of sugar furnishes two of alcohol and two of carbon dioxide. In ordinary practice this complete reaction never occurs, but if by the constant removal of carbon dioxide and alcohol, by means of a mercury pump, atmospheric air be excluded, the equation is nearly realised. M. Pasteur, in his quantitative research on the products of ordinary fermentation,* found that 100 parts of cane-sugar, corresponding to 105*26 parts of gi-ape-sngar, gave nearly Alcohol, ... 51-11 48 '89 according to Gay Lussac's equation. 0"53 excess over ,, ,, Carbon dioxide, Succinic acid, . . 0"67 Glycerin, . . 3-16 Matter united with ) ferments, . . \ 1-00 105-36 * The method adopted by INI. Pasteur to detect and measure quantita- tively the glycerin and succinic acid contained in a fermented liquid, was as follows : — " The liquid, when the fermentation was over, and all the 500 foods: their composition and analysis. [§251. M. Monoyer * has represented Pasteur's results in the form of the following equation : — 4(Ci2H240i2) + GHoO = 2(C4Hc04) + 12(C3H803) + 4CO2 + O2, tlie free oxygen being supposed to serve for the respiration of the yeast cells. Pasteur's equation is far more com])lex, and represents no free oxygen as produced ; yet the fact of fermenta- tion going on briskly in a vacuum, as well as other considerations, points to the jirobable correctness of Monoyer's supposition. As a secondary product, there is also constantly acetic acid. This is generally considered to originate from the ferment itself. With ethyl-alcohol there is also produced, when complex matters are fermented, several other homologous alcohols. For example, if potatoes are fermented, on distilling off the more volatile portions, and collecting separately the final distillate, this is found to consist of jiropyl, butyl, amyl, caproic, ananthyl, and caprylic alcohols, f § 251. Yeast. — Thei-e are no less than three methods of causing the wort of beer to ferment, but none of the thi-ee presents any very distinct variety of the yeast ferment. The one generally employed in England is what is called surface fermentation, in which the starch of the malt is changed into sugar by successive infusions, and the fermentation takes place at from 15° to 18°. This in breweries is done in large open vats ; the yeast floats on the surface, and can be removed by skimming. The second method, more in use in Germany, is fermentation sugar had disaiipearecl (which required from fifteen to twenty days under good conditions), was passed through a filter, accurately weighed against another made of the same paper. After having been dried at 100". the dried deposit of the ferment collected at the bottom of the vessel was also weighed. The filtered liquid was subjected to a very slow evaporation (at the rate of from twelve to twenty hours for each half litre) ; when reduced "to 10 or 20 cc. the evaporation was completed in a dry vacuum. The syrnp obtained was next exhausted with a mixture of alcohol and ether (Tpart alcohol of 30" and To parts ether), which extracted completely all the succinic acid and glycerin. The ether-alcohol was distilled in a retort, then evaporated in a water-batli, and aftei'wards in a dry vacuum. To this was added a little lime water, to fix the succinic acid, and again the mixture was evaporated, aud then the glycerin was separated from the succinate of lime by ether-alcohol, which only dissolved out the glycerin. This ether- alcohol solution on evaporation, which was finished as before in a dry vacuum, left the glycerin in a state fit to be weighed, and the calcium succinate was puriHed by treatment with alcohol of SO per cent., which only dissolved out the foreign matters." — Schiilzenbtrger on '■^Fermentation." Lond., 1876. * " I'hSse de la Faculte de Medecine de Strasbourg." t M. Jeanjean has separated camphyl-alcohol (CioHicO) from the distillate of madder. § 251.] FERMENTED LIQUORS. 501 by sedimentary yeast {Unterhefe). The starch is transformed by decoction, and the temperature (12° to 14°) is much lower than the former. The yeast forms a sediment on the bottom ; and when the first and most active fermentation is over, the clear liquid is run off into proper vessels ; but the yeast not having been aJl deposited, a slow fermentation still goes on for a long time. A third method, employed in Belgium, is to leave the wort to itself, having first placed it \inder proper conditions. The microscopical appearance of yeast is that of a number of round or oval cells, from -00031 to -00035 in. (-008 to -009 mm.) in their greatest diameter. They are transparent, with one or two vacuoles ; contain often a somewhat granular protoplasm, and occur together, united two by two; or, if in active growth, in groups, of which seven is a very common number. These groups are derived from offshoots or buds from a single cell, which will be found somewhere near the middle, and can be identified by its greater size. The usual teaching as to the mode of propaga- tion has been hitherto that this growth is effected by a true budding ; but according to Dr. de Vaureal, the supposed bud- ding is an optical delusion. He considers the utricle of yeast as allied to the spermogones of Tulasne, The nucleolar elements of the cell are spermatic ; and being set free by the rupture of the cell-wall, produce new cells. This explanation would at all events account for the ready manner in which yeast germs are carried about by the air. Schlossberger and many other chemists have determined the chemical composition of yeast. Schlossberger gives the mean of two analyses as follows [the elementary composition after removal of ash] : — Surface Yeast. Sedimentary Yeast. Carbon, 49-9 48 Hydrogen, . . . . 6 6 6 5 Nitrogen, .... 12-1 9-8 Oxygen, .... 31-4 357 [Ash, 2-5 3-5] The ultimate principles contained in yeast appear to be, certain albuminoid substances, tyrosine, leucine [neuclein?], cellulose, and some other hydro-carbons, but the matter is not yet fully worked out. The ash of yeast has been analysed by Mitscherlich, and Schutzenberger arranges his analyses as follows : — 502 FOODS: THEIR COMPOSITION AND ANALYSIS. [§ 252. Phosphoric acid, Potash, ..... Soda, Magnesium Phosphate (M2'o"2P04), Calcium Phosphate (Ca„2P04), . face Yeast. Sediment Yeast. 41-8 39-8 39-5 28-3 16- Tlie researches of Ergol, Rees, Pasteur, and otliers have shown that there are a great variety of alcoholic ferments : the wine ferment is distinct in size and shape from the beer-yeast ferment ; while the ferment of fruit-juice, again, differs in the figure of its cells from either ; and to the various alcoholic ferments names have been given — e.g., Saccharomyces ellipsoideus, S. exiguus, S. conglomeratiis, S. apiculatus, &c. All of them, however, are of quite the same general type as yeast ferment. § 252. "When liquids become sour, ropy, or putrid, in each case the change has been produced by a particular ferment. The lactic acid ferment, e.g., decomposes sugar into lactic acid ; the butyric acid ferment attacks fatty matters, and separates in a free state biityric acid ; and 2}ut7'id ferments, by their aviditv for oxygen, split iip complex organic matters into a vai-iety of substances. Of these ferments, the most important in reference to beer is the lactic acid ferment. In normal alcoholic fermentation, M. Pasteur has proved that thei-e is not the smallest production of lactic acid, and when this does appear, it is certain that it has originated in and contaminated the yeast used. These ferments can be recognised by the microscope, and they should be looked for generally in the sediment. For the purpose of collecting the sediment of beers, wines, water, &C., the author has devised the followingtube (fig. 65). The tube is of the capacity of a litre, and at the lower end is conical and open ; on to this conical end is ground a glass cap, C, which is in poiait of fact a shallow cell about an inch in [ depth ; P is a glass rod, the end of which fits easily into the narrow part of the tube, and is ground so as to make a perfect joint. The use of the tube is as follows : — The liquid under examination is placed in the tube closed by the cap, and the plunger P is removed ; when the sediment has all collected in C, the plunger is very slowly and carefully inserted, so as to stopper the lower end of the tube ; the cap may then be removed for micro- i^C scopical examination. Fig. 65. Fig. 66 is a representation, after Pasteur, of the ferments which characterise sour or turned beer. 503 Fig. 66. 1. Turned beer — filaments simple or articulated into chains of diiferenfc size ; diameter TrirTru- inch. 2. Lactic ferments of beer and wort— small, fine, and contracted in the middle ; diameter a little greater than No. 1. .3. Ferments of putrid wort and beer — mobile filaments generally, which ajtpear at the commencement of fermentation when it is slow ; invari- ably the result of defective working. 4. Ferments of viscous wort and ropy beer, rare. ' 5. Pungent sour beer with acetic odour — chaplets of Mycoderma acefi. 6. A wort deposit. 7. Beer of a peculiar acidity — having a detestable taste and smell, generally found with No 1, but more to be feared than No. 1. BEER. § 253. The most accurate definition of beer,* as brewed in the present day, is that of a fermented saccharine infusion, to which has been added a wholesome bitter. The chief constituents of beer, stouts, and porters, are — (1.) Water containing in solution (according to its origin) various salts. Distilled water is never used in brewing. (2.) Alcohol. (3.) Carbonic and acetic acids. (4.) Malt extract— Malt, sugar, dextrin, albuminous constituents, and ash. * From the decisions in Pashler v. Stevenitt (.35 L.T., 862), and Webb v. Knight (L.Pi,. 2, Q.B.D. 530), it is laid down that beer must be " the article ordinarily sold under that name," and that it "would be to the prejudice of the purchaser to sell him or her as beer an article not of the nature, substance, and quality of that ordinarily sold as such, whether containing ingredients injurious to health or not." (The Sale of Food and Drugs Acts, by W. J. Bell, Esq. London, 1886). 504 FOODS: THEIR COMPOSITION AND ANALYSIS. [§ 253. (5.) Bitter principles, occasionally derived solely from the hop, but very commonly supplemented by so-called " hop substitutes." Samples of the latter examined by the author all contained " quassia," and portions of the following plants were identified— Caluniba, chirata, gentian, and wormwood. (6.) The ash derived from the water, the malt, and the bitters. To these must be added volatile and essential oils, vegetable gelatine, glucinic acid derived from the sugars, lactic acid, and, iu porters, caramel and assamar. The general composition of the chief ingredients of beer may be gathered from the following table, taken from Mr. Watts' Dictionary : — TABLE XXXVI.— SPECIAL EXAMINATION OF CERTAIN BEERS. ri*j _•■"■ 1. t.' o Name of Beer. 1! |« .2 8 SS Analysed by 1^ . - Journ. Pharm., [3], xxi. 26. HusEMANN. — " Die Pflaiizenstoffe." Berliu, 1871. Nativelle.— /oM/v).. Chim. Med., xxi, 69. Scribe. — Compt. Bend., xv. 802. 516 foods: their composition and analysis. [§§263,264. ing about 250° ; it has a bitter taste, -with a weak acid reaction, and is easily soluble in boiling water, with the production of a yellow colour ; it dissolves also in boiling alcohol, but is very little soluble in ether. Daphnin dissolves slightly in cold water, easily in hot ; it is also very soluble in hot alcohol, but is not dissolved at all by ether. In solutions of the caustic and the carbonated alkalies it dissolves with the production of a yellow colour, and is also easily soluble in acetic acid. On boiling with a dilute acid, daphnin breaks up into daphnetin and sugar, and emulsin and fermenta- tion with yeast have a similar effect. An aqueous solution of chloride of iron px'oduces, when cold, a blue colour, and if the liquid is boiled, a dark yellow precipitate. Nitric acid colours it red.* § 263. Gentianin, Cj^HjoOg, discovered by Henry and Caventou in 1821, but first prepared pure by Trommsdorfl", is found, like gentiopicrin, in the root of the Gentiana lutea, L. It forms long, pale-yellow, silky needles, without smell or taste, which may be sublimed above 300° without decomposition. Its solubility, according to Leconte, is as follows : — 1 part requires of cold water 5000 parts, of boiling 3850 parts ; of cold absolute alcohol 455 parts, of boiling 62-5 parts ; of cold ether 2000 parts, for solution. Concentrated sulphuric acid dissolves it with a yellow colour; on dilution with water it separates unchanged; on being boiled with dilute sulphuric acid there is no change. If treated with pure nitric acid (1-43 specific gravity), a dark-green solution is obtained, and on adding water carefully dinitro- gentianin, Cj4Hg(N02)205 + HoO separates out as a green powdei*. If similarly treated with strong nitric acid, and subsequent addi- tion of water to the solution, yellow microscopical prisms are separated, probably trinitro-gentianin. Gentianin reduces nitrate of silver. •{• § 264. Gentiopicrin, Q^^K^fP^^, first prepared pure by Kromayer in 1862, is a glucoside found in the fresh root of the Gentiana lutea. It crystallises in colourless needles, with one atom of water * Bibliography. Gmelix and Bar. — Schweigcj. Journ., xxxv. 1. HusEMANN. — "Die Pflaiizenstoffe." Berlin, 1871. EocHLEDER. —/oMrw. Pract. Chem., xc. 442. Vauquelin.— ^H?j. C/((m., Ixxxiv. 174. ZwENGER. — Ann. Chem. Pharm., cxv. 1. t Baumert.— ^ww. Chem. Pharm., Ixii. 106. Henry and Caventou. — Journ. Pharm. [2], vij. 173. HusEMANN.— "Die Pflanzenstoife." Berlin 1871. Leconte.— Jowrw. Pharm., [2], xxiii. 465. Tkommsdorff.— .4«?i. Chem. Pharm., xxi. 134. §§265,266.] beer: other bitter principles employed. 517 of crystallisation. The crystals effloresce in the air, and lose their transparency, becoming white and opaque. The anhydrous crystals melt between 120° and 125° to a brown fluid, which coagulates amorphously, and at higher temperatures is fully destroyed. Water dissolves it easily, but it is insoluble iu absolute alcohol and in ether, though, on the other hand, weak alcohol is an excellent solvent for it. One of the best tests for its presence is the action of concentrated sulphuric acid, giving in the cold a colourless solution, but producing with slight warm- ing a carmine-red colour, and precipitating on the addition of water in grey flocks. It reduces an ammoniacal solution of silver nitrate, and in boiling with dilute sulphuric acid splits up into sugar and gentiogeniu. Gentiogenin, Cj^HjgOg, is an amorphous, yellow-brown powder, of neutral reaction and bitter taste, not easily soluble in cold water, but dissolving readily in alcohol and ether.* § 265. Menyanthin, CggH^gO^, a glucoside obtained piire by Ludwig and Kroraayer in 1861, from the Menyanthes trifoliata, L. As hitherto prepared, menyanthin is an amoi-phous, terebin- thinate mass, becoming slowly solid on drying over sulphuric acid. It has a neutral reaction, and its taste is strongly and purely bitter. It softens at 60° to 65°, and melts at 10° to 15° to a thin, clear fluid, which again solidifies to a hard ti-ansparent mass ; by stronger heating it is entirely destroyed. Concen- trated sulphviric acid gives with it a yellow-brown colour, which on standing becomes violet-red, and grey flocks are separated on the addition of water. By heating with dilute sulphuric acid it splits vip into sugar and menyanthol. Menyanthol is an oil having an acid reaction, and an odour like that of oil of almonds ; it is changed by the air (as well as by melting with potash) into a crystalline acid capable of being sublimed. t § 266. Quassiin, C^oH-^^^Og, a bitter principle, discovered in 1835 by Winckler, in the bark of the Quassia aniara, L., and Picroina, excelsa, L. It forms white, opaque, glittering crystals, without odour, and of extremely bitter taste. On heating, it melts and solidifies as a transparent yellow mass; at decomposition tem- peratures it burns like resin, if exposed to the air. Tannic acid precipitates it from an alcoholic solution in thick, white flocks ; * Bibliofjraphy. HusEMANN.— "Die Pflanzenstoffe." Kromayee. — Arch. Pharm. [2], ex. 27. t HusEMANN. — Die Pflanzenstoffe. Berlin, 1S71. Kromayee. — Arch. Pharm. [2], cxxiv. 37. Ludwig and Kromayee. — Arch. Pharm. [2], cviii. 263. 518 FOODS : THEIR COMPOSITION AND ANALYSIS. [§§ 267, 268. cold concentrated sulphuric acid dissolves it, without the pro- duction of colour ; and on dilution with water it is apparently jirecipitated without change."' § 267, The Ash of Beer. — The ash of beer contains the mineral constituents that previously existed partly in the water, partly in the hop, and j^artly in the malt used. It would appear that the ferric oxide, a certain proportion of phosphoric acid, a small portion of the lime and magnesia, with a great part of the silica, remain undissolved, and do not p-jss i_nto tho beer; the rest are dissolved. The following table gives the average composition of the beer «.sh of commerce : — Beer Ash. Potash, 37-22 Soda 8-04 Lime, . . . . . . . . 1-93 Magnesia, 5 'SI Iron oxide, traces Suiphiu-ic acid, 1-44 Phosphoric acid, 32-09 Chlorine, 2-91 SiUca, 10-S2 The table on next page may be also useful, showing analyses of ash by Walz and Dickson (" Dictionary of Chemistry, Ax-ts, and Manufactures," edited by Vincent). § 268. Analysis of Beer. — The ordinary full analysis of beer determines — (1.) The alcohol (2.) The carbonic acid. (3.) The volatile and fixed acids. (4.) The percentage of malt extract, and, if necessary, its com- j^osition. (5.) The hop resin and glycerin. (6.) The nature of the bitter used. (7.) The general composition of the ash, and especially the « amount of chlorine. (1.) The alcohol is found most accurately by the distillation process described at p. 476, but it is often determined in the following manner : — Shake up the beer in a flask, so as to * BibHography. HusEMANN. — "Die Pflanzenstoffe." Berlin, 1S71. WiGGERS.— .d?m. Che7n. Pharm., xxi. 40. WiNCKLEK.— iiV/Jcri!. Pkarm., xxi. 40. § 268.] ANALYSIS OF BEEE. 519 i^ <» p •^^ "» p ■* •* •* p 00 Is o ^ ^ ,—1 b ,1h 00 00 fSg- ca '-' - s 1 1 1 1 1 1 1 1 ll <^ <=p 1 CO c^ > ■ t:' p II 00 t^ (N CO ,_, plicable to all liquids from which carbonic anhydride can be expelled by boiling. (3.) Volatile and Fixed Acids. — The acetic acid is obtained by distilling the beer nearly to dryness, and estimating the acidity of the filtrate by decinormal solution of soda. Should the residue in the flask or retort be still acid, a little water should be added, and the distillation again continued to dryness ; any acid now remaining is certain to be a fixed acid, probably lactic. It may be estimated by titration, and returned as lactic. The equiva- lent of anhydrous lactic acid is 90 ; hence 1 cc. of d. n. soda = 9 mgrms. of lactic acid. Should it be specially necessary to deter- mine the percentage of lactic acid, a sufficient quantity of beer — say oOO cc. — is taken, evaporated to a small bulk, diluted with ■water, filtered, and mixed with a little sulphuric acid ; pure carbonate of baryta is now added, and the whole warmed on the water-bath for some time. The liquid is then freed from the precipitate of sulphate of baryta by filtration, and the pre- § 268.] ANALYSIS OF BEER. 521 cipitate well washed with hot water. This filtrate is evaporated to a syrup, and treated, when cold, in a tube or separating- funnel with a mixture of one part of sulphui'ic acid, one of water, one of alcohol, and ten of ether; the ethereal layer is separated in the usual way, and evaporated. The lactic acid thus obtained is still impure, and it is best to dissolve in water, saturate with freshly-precipitated carbonate of zinc, and estimate as zinc lactate, the latter containing 54--i9 per cent, of anhydrous lactic acid. In most cases, however, the error will not be great, if the total acidity of the beer is taken directly without distilla- tion, and returned as acetic acid. (4.) The Malt Extract. — The extract of beer can be determined by evaporating down a carefully measured quantity, and weigh- ing the dry residue. In order to do this with any approach to accuracy, the smallest possible quantity should be taken — 5 cc. or 5 grms. is quite sufficient. This small quantity, spread out as a thin film on the bottom of a tolerably capacious platinum dish, can be thoroughly dried over the water-bath in two or three hours, while if such quantities as 25 cc, 50 cc, or 100 cc. are taken, to get the extract completely dry is very tedious, and usually requires a higher temperature than 100°. It is, however, found in practice much more convenient to dispense with this drying altogether, the alcohol and carbonic acid being driven off, as before described, the beer made up to its first bulk, the specific gravity taken, and the amount of malt extract determined by the aid of the following tables. If the beer has been distilled, the residue in the retort or flask can be made up to the original bulk, brought to the proper temperature, and treated as just described. The alcoholic strength, the acetic acid, and the amount of malt extract being known, the analyst can now give a fairly approximate estimate of the amount of malt originally used in the brewing of the beer. To do this it is necessary to calculate the " original gravity " of the beer. The specific gravity of the alcoholic distillate (oi*, if an indirect process has been used, .the specific gravity of the alcoholic strength) subtracted from 1000, gives a number called the " spirit indication." The degrees of gravity lost ai'e then ascertained by the aid of the following tables, using the first if the beer has been distilled, and the second if the evaporating process has been used. The degrees of gravity thus found are added to the specific gravity of the boiled beer, and the number thus obtained is called " the original gravity of the wort." On reference to Table XL., the amount of malt extract is determined, which cori-esponds to this original gravity. 522 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 268. TABLE XL.— Specific Gravity and Strength of Malt Extract. Specific Malt SpeciQc Malt Specific Malt Specific Malt Gravity. Extract. Gravity. Extract. Gravity. Extract. Gravity. Extract. 1,0000 0,000 1,0047 1,175 1,0094 2,350 1,0141 3,525 1,0001 025 48 200 95 375 142 550 2 050 49 225 96 400 143 575 3 075 1,0050 250 97 425 144 600 4 100 51 275 98 450 145 625 5 125 52 300 99 475 146 650 6 150 53 325 1,0100 500 147 675 7 175 54 350 101 525 148 700 8 200 55 375 102 550 149 725 9 225 56 400 103 575 1,0150 750 1,0010 250 57 -425 104 600 151 775 11 275 58 450 105 625 152 800 12 300 59 475 106 650 153 825 13 325 1,0060 500 107 675 154 850 14 350 61 525 108 700 155 875 15 375 62 550 109 725 156 900 16 400 63 575 1,0110 750 157 925 17 425 64 600 111 775 158 950 IS 450 65 625 112 800 159 975 19 475 66 650 113 825 1,0160 4,000 1,0020 500 67 675 114 850 161 025 21 525 68 700 115 875 162 050 22 550 69 725 116 900 163 075 23 575 1,0070 750 117 925 164 100 24 600 71 775 118 950 165 125 25 625 72 800 119 975 166 150 26 650 73 825 1,0120 3,000 167 175 27 675 74 850 121 025 168 200 28 700 75 875 122 050 169 225 29 725 76 900 123 075 1,0170 250 1,0030 750 77 925 124 100 171 275 31 775 78 950 125 125 172 300 32 800 79 975 126 150 173 325 33 825 1,0080 2,000 127 175 174 350 34 850 81 025 128 200 175 375 35 875 82 050 129 225 176 400 36 900 83 075 1,0130 250 177 425 37 925 84 100 131 275 178 450 38 950 85 125 132 300 179 475 39 975 86 150 133 325 1,0180 500 1,0040 1,000 87 175 134 350 181 625 41 025 88 200 135 375 182 550 42 050 89 225 136 400 183 575 43 075 1,0090 250 137 425 184 600 44 100 91 275 138 450 185 625 45 125 92 300 139 475 186 650 46 150 93 325 1,0140 500 187 675 m. SPECIFIC GRAVITY OF MALT EXTRACT. TABLE XL.— Continued. 523 Specific Malt Specific Malt Specific Malt Specific Malt Gravity. Extract. Gravity. Extract. Gravity. Extract. Gravity. Extract. 1,01S8 4,700 1,0235 5,875 1,0282 7,024 1,0329 8,170 189 725 236 900 283 048 1,0330 195 1,0190 750 237 925 284 073 331 219 191 775 238 950 285 097 332 244 192 800 239 975 286 122 333 260 193 825 1,0240 6,000 287 146 334 292 194 850 241 024 288 170 335 316 195 875 242 048 289 195 336 341 196 900 243 073 1,0290 219 337 365 197 925 244 097 291 244 338 389 198 950 245 122 292 268 339 413 199 975 246 146 293 292 1,0340 438 1,0200 5,000 247 170 294 316 341 463 201 025 248 195 295 341 342 488 202 050 249 219 296 365 343 512 203 075 1,0250 244 297 389 344 536 204 100 251 268 298 413 345 560 205 125 252 292 299 438 346 584 206 150 253 316 1,0300 463 347 609 207 175 254 341 301 488 348 633 208 200 255 365 302 512 349 657 209 225 256 389 303 536 1,0350 681 1,0210 250 257 413 304 560 351 706 211 275 258 438 305 584 352 731 212 300 259 463 306 609 353 756 213 325 1,0260 488 307 633 354 780 214 350 261 512 308 657 355 804 215 375 262 536 309 681 356 028 216 400 263 560 1,0310 706 357 853 217 425 264 ■ 584 311 731 358 877 218 450 265 609 312 756 359 901 219 475 266 633 313 780 1,0360 925 1,0220 500 267 657 314 804 361 950 221 525 268 681 315 828 362 975 222 550 269 706 316 853 363 9,000 223 575 1,0270 731 317 877 364 024 224 600 271 756 318 901 365 048 225 625 272 780 319 925 366 073 226 650 273 804 1,0320 950 367 097 227 675 274 828 321 975 368 122 228 700 275 853 322 8,000 369 146 229 725 276 877 323 024 1,0370 170 1,0230 750 277 901 324 048 371 195 231 775 278 925 325 073 372 219 232 800 279 950 326 097 373 244 233 825 1,0280 975 327 122 374 268 234 850 281 7,000 328 146 375 292 624 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 268, TABLE XL.— Continued. Specific Malt Specific Malt Specific Malt Specific Malt Gravity. Extract. Gravity. Extract. Gravity. Extract. Gravity. Extract 1,0.376 9,316 1,0423 10,452 1,0470 11,571 1,0517 12,690 377 341 424 476 471 595 518 714 378 365 425 500 472 619 519 738 379 389 426 523 473 642 1,0520 761 1,0380 413 427 547 474 666 521 785 381 438 428 571 475 690 522 809 382 463 429 595 476 714 523 833 383 488 1,0430 619 477 738 524 857 384 512 431 642 478 761 525 881 385 536 432 666 479 785 526 904 386 560 433 690 1,0480 809 527 928 387 584 434 714 481 833 528 932 388 609 435 738 482 857 529 976 389 633 436 761 483 881 1,0530 13,000 1,0390 657 437 785 484 904 531 023 391 681 438 809 485 928 532 047 392 706 4.39 833 486 952 533 071 393 731 1,0440 857 487 986 534 095 , 394 756 441 881 488 12,000 535 119 395 780 442 904 489 023 536 142 396 804 443 928 1,0490 047 537 166 397 828 444 952 491 071 538 190 398 853 445 976 492 095 539 214 399 877 446 11,000 493 119 1,0540 238 1,0400 901 447 023 494 142 541 261 401 925 448 047 495 166 542 285 402 950 449 071 490 190 543 309 403 975 1,0450 095 497 214 544 333 404 10,000 451 119 498 238 545 357 405 023 452 142 499 261 546 381 406 047 453 166 1,0500 285 547 405 407 071 454 190 501 309 548 428 408 095 455 214 502 333 549 452 409 119 456 238 503 357 1,0550 476 1,0410 142 457 261 504 381 551 500 411 166 458 285 505 404 552 523 412 190 459 309 506 428 553 547 413 214 1,0460 333 507 452 554 571 414 238 461 357 508 476 555 595 415 261 462 381 509 500 556 619 416 285 463 404 1,0510 523 557 642 417 309 464 428 511 547 558 666 418 .333 465 452 512 571 559 690 419 357 466 476 513 595 1,0560 714 1,0420 381 467 500 514 619 561 738 421 404 468 523 515 642 562 761 422 428 469 547 516 666 563 785 § 268.] SPECIFIC GRAVITY OF MALT EXTRACT. 525 TABLE XL.— Continued. Specific Malt Specific Malt Specific Malt Specific Malt Gravity. Extract. Gravity. Extract. Gravity. Extract. Gravity. Extract. 1,0564 13,809 1,0604 14,761 1,0643 15,674 1.0682 10,581 565 833 605 785 644 697 ■ 683 604 566 857 606 809 645 721 684 627 567 881 607 833 646 744 685 650 568 904 608 857 647 767 686 674 569 928 609 881 648 790 687 697 1,0570 952 1,0610 904 649 814 688 721 571 976 611 928 1,0650 837 689 744 572 14,000 612 952 651 860 1,0690 767 573 023 613 976 652 883 691 790 574 047 614 15,000 653 907 692 814 575 071 615 023 654 930 693 837 576 095 616 046 655 9.J3 694 860 577 119 617 070 656 976 695 883 578 142 618 093 657 16,000 696 907 579 166 619 116 658 023 697 930 1,0580 190 1,0620 139 659 046 698 953 581 214 621 162 1,0660 070 699 976 582 238 622 186 661 093 1,0700 17,000 583 261 623 209 662 116 701 022 584 285 624 232 663 139 702 045 585 309 625 255 664 162 703 067 586 333 626 278 665 186 704 090 587 357 627 302 666 209 705 113 588 381 628 325 667 232 706 136 589 404 629 348 668 255 707 158 1,0590 428 1,0630 371 669 278 708 181 591 452 631 395 1,0670 302 709 204 592 476 632 418 671 325 1,0710 227 593 500 633 441 672 348 711 250 594 523 634 464 673 371 712 272 595 547 635 488 674 395 713 295 596 571 636 511 675 418 714 318 597 595 637 534 676 441 715 340 598 619 638 557 677 464 716 363 599 642 639 581 678 488 717 386 1,0600 666 1,0640 604 679 511 718 409 601 690 641 627 1,0680 534 719 431 602 714 642 650 681 557 1,0720 17,454 603 738 526 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 268. 1— ipHp-<(M(MCOCO'^-*OiOOCO CSOCKMCO-HOOt^piMpOO'^qOCO Q i iOTf(OD(Ml^-r-OC-(/-H*l:^OOC5'-Hio pi^T*o o -* o LO b in o o rt rt J) C) (H CO Tt< ■* lO lO o o COCOOi-iiCOSOIMCOt-t^OCJO-t* OOpi;^'7Hi07H«»t--^CqpC0p00ip bcoooinc3TtioocoG5Tj r-i .-< p— I pcooopiro>pp-*p . rti-(i-((N(NCi5C«5'*'^iOiOOl.- > Tt^(^lfO•rt^qo■5J^pa)M>p^-p-*p'i^ . CO 7-qopprtipipcpt--pc^ippc^cr) lO ^-Tt^»ppp»ppcp'7^»pppcpQ0C^ "* -^p'THC'i'^'T-occpp-THipcoirit^ 1— 1 lo Ci cc t^ (M o — 1 o c-i t^ oi t^ ^ oo : rt r-l Ol C4 CO CO Tl^ Tt< o o o o CO ppt^-coppTHOpipiOCicor^— 1 ^r-i(Mc^coco^TtHicir50 cq r-C^(J1rJ ■—1 -.^ '^ ^ 9 . § »o CO CO CO H-l -H H as 02 in o § ?5 n ^ C5 t^ ^ o> § ?J r- =? ^ O S u 5 ^^ r- a '^ "-* '"' s CO 9 O CO ,_, CO <£> o ,_, ■^< t-~. 05 iCl CO -* to oo Pm ,_! ,_ ,_) 02 M o O 9 "* t- Oi (M >o t^ o CO lO Oj CO O . 465. Rauwez, J. V. — Moyen simple de reconnaitre une bi&re k I'aloes. Journ. Chim. Med., 4e serie, 1862, t. x., p. 233. Schmidt. — Bifere falsifiee avec de la picrotoxine. Journ. Pharm. et Chim,, 3e serie, 1863, t. xliii., p. 170. ScHiJTZENBERGER, P. — " On Fermentation." London, 1876. Schulze.— Maltose. Deut. Chem. Ges. Ber., vij. 1047-1049. SouBEiRAN. — " Dictionnaire des falsifications." Paris, 1874. Stierlin, R.— Das Bier und seine Verfiilschungen, Bern, 1878. Sullivan, C. O. — Maltose. Journ. Chem. Soc, i., 1876, p. 478. Vincent. — " Dictionary of Arts and Manufactures," Art. Beer. VoGEL, Herm. W. — " Practische Spectral Analyse," &c. Nordlingen, 1877. Wanklyn in " Manual of Public Health," ed. by Ernest Hart. Watts. — " Chemical Dictionary." Wenke. — " Das Bier und seine Verfillschung." Weimar, 1861. Wittstein, G. C.— Detection of Adulterations in Beer. Arch. Pharm. [3], V. 25-23. § 274.] 545 WINE. § 274. Constituents of Wine. — Wine is the fermented juice of the grape, with such additions only as are essential to the stability, or keeping of the liquid {Dupre). The constituents of grape juice and wine may be arranged and compared as follows : — Water. Grape sugar. Must. Albuminoid bodies. Wtne. Water. Grape sugar (0 to several per cents.) Alcohols (mainly ethylic, but also small quantities of the higher alcohols, such as propylic, butylic, amylic, and others). Residues of albuminoid bodies. Aldehydes (mainly ethylic). Isobutylglycol.* Acetal. Furfurol. Acetic acid. Succinic acid. Hydro-potassic tartrate. Tartrate of lime. Gum. Malic acid (in bad seasons). /Tartrate of lime, (In smaller propor- \ Tartaric acid, ttions thanin"Must." Vegetable mucus. Gum. Malic acid (in bad seasons). Salts of ammonia and of similar! Small mixtures of colouring-matters. Colouring-matters. Glycerin. Organic acids in combination and Organic matters in combination and certain extractive matters. certain extractive matters. Esters — Acetic, caproic, caprylic, butyric, and tartaric esters have been identified. Tannin. Mineral matters. t Mineral matters. A few ferment cells and similar forms. * A. Henninger, operating on 50 litres of Bordeaux, succeeded in isolating, by fractional distillation, 6 grnis. of isobutylglycol, boiling point 178°-5. Making a correction for the glycol carried away by aqueous vapour, he considers the amount in wine to be equal to 0'05 per cent., or about one-fifteenth of the glycerin. {Comptes Rendus, xcv., 94-96.) t Among the mineral matters of the grape, and, therefore, generally also in wines, is a small quantity of boric acid. The mineral matters of the ash of eighteen samples of grape juice have been quantitatively determined by Mr. Carter Bell, Analyst (November, 1881) ; the chief results are as follows : — o a i r-.i i o a 1 a o< P s 1 It So. 3 g It i (2 ^ 3 ^ ^ s s s ^5 Ph i« Maximum, . .';4-24 10 -.54 3-39 13-08 7-28 13-99 12-57 1-OS 3-S5 '>3-7.S •98 Minimum, . 31-23 -29 -29 3-14 •09 -0.5 •90 -05 •05 3-13 •03 Mean, . . 42-14 3-37 1-09 9-14 3-00 4-55 y-07 •03 •87 12 •78 •29 This may be compared with the mineral constituents of wine given at p. 579. 36 546 FOODS : THEIR COMPOSITION AND ANALYSIS. [§§ 275, 276. § 275. Changes taking place in Wine through Age. — Berthelot * lias made several analyses of wines 100 and 45 years old respec- tively, which are interesting as contributing to more accurate knowledge regarding the effect of age upon wine. The wines were both samples of Port. The one 100 years old had a large deposit of colouring-matter, and was yellow ; the colour of the second sample was dark, but yet lighter than that of new wine. The results of the analyses are as follows : — Specific gravity (at 10°) Total residue at 100°, .... Sugars, reducing, Sugars, after the action of dilute acid. Acid, calculated as tartaric acid, grms. per litre, . . , Tartaric ether, ...... Cream of tartar, Alcohol, per cent., ..... Sugar, reducing, Cane-sugar, Pure acids, Acids as ethers, Cream of tartar. Glycerin and other matters, This research of Berthelot's, as well as the more recent investi- gations of Schmidt,! show that there is a gradual deposit of the colouring-matter, and that some of the sugar has disappeared from the old wine, which gives a smaller residue. Cane-sugar is practically absent in the sample 100 yeai's old, a fact which Berthelot interprets as confirmatory of his observation of the slow invertive action of inorganic acids on cane-sugar. The alcohol is lower in old wines than new, and the acidity tends to diminish, the acids combining with alcohols to produce esters. The experiments of Macagno may also be here cited, from which it appears that in wines of the same class the tannin decreases through age, while the glycerin increases. § 276. Adulterations of Wine. — The adulterations of wine are as folIoAvs : — Watering, fortifying with spirit, fortifying and watering, the addition of various fermented liquids (such as * Comptes Bendus, 88, 1879, 626. t Die Weine. Berlin, 1893. Port wine, Port wine. 100 years old. 45 years old. •988 •991 3-36 5 -.30 1-25 315 1-29 3-68 5-17 5-46 111 1-17 •27 •42 15-9 16^1 llowing results : — Wine Wine 100 years old. 45 years old. 125 3^15 •04 •53 •51 •52 •27 •28 •03 •04 2-10 4-52 116 •98 § 277.] ANALYSIS OF WINE. 547 wines of low value to those of high value — that is to say, alco- holic liquids made from the fermentation of glucoses or various sugars, or wines made from raisins or figs to wines made from the grape — the mixing or blending of wines (this may be a necessary operation in some cases, in others it takes the form of sophistication, when wines of higher quality are mixed with wines of low quality and sold as wines of the higher quality), plastering, the addition of bitartrate of potash and ethers (such as osnanthic ether) to give a fictitious appearance of age, the addition of alum to brighten the colour, the artificial colouring of wines, the addition of antiseptics (such as salicylic acid), and the addition of fluoborates or fluosilicates. § 277. The Analysis of Wine. — The analysis may be divided conveniently into : — I., Physical characters ; II., the estimation and qualitative detection of constituents volatile at or below 100°; III., estimation and identification of matters not volatile at 100°; IV., the estimation of the total constituents of the mineral matters of the ash, and, similarly, the identification and estima- tion of the separate constituents of the ash. I. Physical Characters. — This embraces the specific gravity and the action on polarised light. II. Constituents Volatile at or heloiv 100°. — This embraces the estimation of alcohol, volatile esters, sulphurous acid, aldehyde- sulphurous acid, volatile acid, and, as far as practicable, the identification of the constituents comprised under those names. III. Constituents not Volatile at or below 100°. — This embraces total extract,* total fixed acid, free tartaric acid, sugars, potassic tartrate, fixed esters, glycerin, potassic sulphate, tannin, and colouring matters. IV. The Ash. — Special processes may be required for the detec- tion and estimation of some of the adulterants mentioned. The table on p. 550 gives, according to this plan, the results of the analysis of a number of wines of considerable age and high price analysed by Dr. Conrad Schmidt, f The tables on pp. 548 and 549 give a number of analyses by Dr. Dupre. * E. Pdegler {Zeit. f. anal. Chemie, 189B, H. 1, s. 27) lias published an easy and rapid method of determining the amount of alcohol and extract in wine. The refraction is taken of the wine by means of an Abbe's or a Pulfrich's refractometer ; the alcohol is next got rid of by boiling a mea- sured quantity ; after cooling and making up with water to the original bulk the refraction is again taken. He finds that a gramme of extract in 100 cc. of wine raises the refraction beyond that of water '00145, and 1 grm. of alcohol in 100 cc. of wine raises the refraction 0-00068. If N -•- the refraction of the undistilled wine ; a, the refraction of the distilled pure water ; a + b, the refraction of the wine freed from alcohol and made rp to the original volume, then — "' - alcohol in grms. per 100 cc, of wine and ^ .,,, .^ extract in 100 cc. wine. 00140 t "Die Weine des herzoglich nassauischen Cabinetskeller," von Hofrath Dr. Conrad Schmidt. Berlin, 1893. 548 FOODS : THEIR COMPOSITION AND ANALYSIS. TABLE XLIV. — Weight in Grajnimes of some of the Chief Undermentioned 1 Particulars of Wines Analysed. 1 o •3 02 < 1 1 < << n § k If f C3 << g.2 SI 3 < 1 H ■3 Doz. Vintage. Hock, white, 30s. 1862 993-43 95-6 3-48 0-57 4-20 40s. 1859 993-48 92-0 4-20 1-14 5-62 2-550 1203. 1857 992-81 104-4 4-31 0-93 5-37 0-675 Claret, . 15s. 1865 995-58 85-3 4-24 1-47 6-08 0-675 4Ss. 1865 995-03 120-0 4-24 1-74 6-41 1-875 66s. 1861 994-73 85-3 3-23 1-80 5-48 1-838 Hungarian, red, 21s. 992-07 113-6 3-56 2-49 6-68 0-600 „ white, 343. 992-88 95-4 5-33 1-47 7-16 0-675 „ 42s. 993 09 94 9 4-74 1-80 6 99 0-375 Greek wine,,, 20s. €94-56 107-2 3-41 3-00 7-16 0-675 „ „ 28s. 992-25 124-5 4-54 1-68 6-64 ... „ „ 363. 993-17 138-9 2-33 1-77 4-54 0-300 Sherry, . 22s. 1865 994-09 172 2-70 1-53 4-61 0-187 ,, high price, 1860 997-93 178-1 3 08 1-68 5-18 0-262 .. „ .1 1857 998-30 184 2-81 162 4-84 0150 Madeira, E.I., 60s. ... 993-94 177-5 3-26 1-68 5-36 0-300 ,, high price, 1812 994-15 180 4-20 3-27 8-25 Port, . 32s. 1864 1004-76 185 3 08 0-84 4-13 0-225 ,, high price, 1854 997-42 175-3 3-54 1-07 4-88 0-225 >» ». ). 1842 986-95 182-6 2-66 1-08 4-01 0-150 Marsala, . 163. old 996-65 167-1 1-88 1-11 3-26 203. ^ 'ery old 999-65 168-9 2-25 1-38 3-98 0-150 CONSTITUENTS OF WINE. 549 Constituents of 1 Litre (1000 cc. or 1000 Volumes, of the Wines (Dupre). •a a H 1 11 4 < i H ii 1 II 1 < < < £ -32 -1 < t ^1 ill 11% .i J= — III 18-63 1-95 0-58 0-76 0-60 0-32 0-132 0-230 0-362 0-360 100-5 18-55 0-12 1-70 07 0-78 085 30 0-199 0-239 0-438 458 95-7 20 GO 1-12 145 0-14 0-46 0-85 0-35 0-225 0-239 0-464 0-493 94-3 21-40 4-31 2-08 60 0-95 0-48 0-33 0-155 0-197 0-352 0-476 74-0 24-33 2-04 2-25 0-66 1-05 0-55 30 0-186 0-248 0-430 0-581 74-6 18-00 95 200 0-38 0-99 0-63 30 0-166 0-216 0-382 0-4-29 88-8 20-85 1-47 1-85 0-41 0-91 0-53 0-35 0-151 0-358 0-509 0-656 77-6 18-20 061 1-75 0-14 0-81 0-80 0-25 0-186 0-271 0-457 0-013 74-5 18-13 0-24 1-88 0-12 0-90 0-85 0-25 0-162 0-273 0-435 0-596 73-0 25-30 2-00 2-25 0-07 1-18 100 0-25 0-224 0-214 0-438 0-690 63-6 24-42 1-12 3-05 0-41 2-01 0-62 25 0-384 0-179 0-563 0-707 79-6 25-50 3-64 3-75 0-21 2-49 1-05 0-45 0-245 1-207 0-453 0-530 85-1 42-00 25-65 4-50 0-07 3-63 080 0-18 0-206 0-216 0-422 0-639 6S-1 53-50 29-70 5-50 0-18 4-41 095 25 0-290 0-391 0-681 0.749 90-8 50-44 35-10 5-13 0-07 4-18 0-88 0-13 0-262 0-469 0-731 0-722 101-2 43-47 20 80 3-90 0-27 2-52 MO 0-42 0-305 0-382 0-687 0-774 88-7 45-41 16-29 359 0-17 1-93 1-49 0-50 0-460 0-773 1-233 1-207 102-1 75-57 43-31 2-48 0-48 1-34 0-65 0-35 0-302 0-128 0-430 0-620 69-4 53-90 22-84 2-58 0-66 1-37 0-55 0-33 0-351 0-220 0-571 0-697 84-9 31-01 10-10 2-10 0-69 0-86 0-45 0-33 0-283 0-331 0-614 0-595 103-2 4'9-83 32-40 2 25 0-21 1-54 0-50 018 0-256 0-189 0-445 0-447 99-3 57-48 37-60 3-13 0-55 1-92 0-65 0-23 0-333 0-216 0-549 0-550 99-8 550 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 277. TABLE XLIVa. — Giving the Chief Results of Schmidt's Analyses (1888-1892) OF certain Fine awd Exceptionally Valuable Old Wines. Parts by weight per Litre. Physical characters. Specific gravity. Min., .... Max., .... Mean, .... Polarisation— 200 mm. tube. (Expressed in Wilds degrees.) Min Max., Mean, Constituents volatile below 100° C. Alcohol. Min., Max., Mean, Volatile Esters. (Expressed in cc. of d. n. KHO.) Min., Max., Mean, Aldehyde-sulphurous acid, (In grms. of S0«.) Min., Max., Mean, Volatile acid. (Expressed as acetic acid.) Min., Max., Mean, Constituents not volatile at 100° C. Extract. Min., .... Max., .... Mean, . . Total tartaric acid. Min., .... Max Mean, .... Free tartaric acid. Min., .... Max., .... Mean, .... Total fixed acid reckoned as tartaric acid. Min., . Max., .... Mean, .... Hoch- aimer, 9 samples. (1706-1S68) 099S1 1 -0044 10105 + 0-16° + 0-49° + 0-33" 37-6(1706) 76-7 56-3 27-6 50-4 32-6 0-059 0-223 0125 l-Ol 2-32 1-44 25-43 27-79 27 50 2-31 3-96 2-72 0-63 1 -.35 84 2-97 5-94 4-62 stein- Marko- berger. brunner, 19 samples 7 sample: (lSll-1873) (1822-186S) 0-99S2 1 •()0-25 0-9990 + 0-05° + 1 -50° + 0-39° 43-7 92-9 67-7 17-2 48-0 32-7 069 0-245 0-170 1-10 2-42 •52 20-98 42-31 29-5 1-74 3-32 2 51 051 102 71 4-10 6-44 4-62 0-9983 1-0018 0-9997 + 0-30° -H-.32° + 0-59° 49-4 90-0 74-4 15-6 49-2 34-4 059 0-214 0-172 1-20 2-15 1-67 23-48 44-06 3178 2-03 3-09 2-26 0-.39 0-72 067 4-13 5-16 4-78 Paides- heimer. 11 samples (1831-1880) 0-9963 1 -0026 0-9995 -f 0-05° + 1 -55° + 45° 46-9 93-6 71-7 27-2 50-4 36-8 0-077 0-260 160 1-04 2-29 1-47 21-80 47-37 30-54 1-80 3-15 2 46 0-39 1-41 0-71 3-80 5-40 4-67 § 277.] ANALYSIS OF WINE. TABLE XLlVa.—Confniued. 551 Hoch- eimer. 9 samples (1706-186J Stein- berger. 19 samples (1811-1873) Marko- brunner. 7 samples. Eudes- heimer. 11 samples (1831-r— Tannin and colouring matters. Min., Max., Mean, Fixed ester. ( Expressed in cc. of d.n. KHO. Min., Max., Mean, Glycerin. Min., Max., Mean, Potassic sulphate. Min., Max., Mean, Ash. Min., Max Mean, 0-19 0-71 0-39 134-1 257-6 188-7 13-06 15-81 12-62 0-76 1-63 115 2-19 2-67 2 55 0-22 0-45 0-36 144-4 488-8 233 9 10-51 18-76 14-64 0-59 1-19 1-78 2-59 2-24 0-23 0-68 0-42 128-4 446-4 268-5 11-70 22-55 16-29 0-44 1-52 0-78 2-13 2-45 226 0-16 0-66 0-28 118-8 436-0 2318 9-79 24-46 14 63 0-47 1-20 0-73 1-87 2-60 2-23 Numerical Eelations of certai> Constituents. Glycerin : Alcohol. (Alcohol = 100.) Min., 15-9 16-8 18-3 14-1 Max., 30-7 32-3 26-1 29-5 Mean, 23 5 215 21-5 20-5 Volatile acid : Total acid. (Total acid = 100.) Min., 15-6 12-6 18-2 16-7 Max., 34-8 38-4 31-6 31-3 Mean, 21-7 23-4 25-8 22-2 Volatile ester to alcohol. (Alcohol = lUO.) Min., 52-9 .33-5 17-4 41-6 Max., 64-7 61-7 69-6 64-0 Mean, 56-3 48-7 49-6 52 3 Volatile ester to volatile acid. (Volatile acid = 1.) Min., 18-2 15-0 9-5 21-9 Max. 26-3 30-0 29-5 34-0 Mean, 22-8 22-2 209 257 Total ester to total acid. (Total acid = 1.) Min., 21-5 29-9 24-6 25-5 Max 46-2 73-3 63-3 69-2 Mean, 33-3 41-4 41-7 40-6 552 FOODS : their composition and analysis. [§ 277. I. Physical Characters. The specific gravity and polarimetric estimations are made on similar lines to those alieady detailed under "alcohol" or •' sugar." II. Constituents Volatile at or Below 100° C* Alcohol. — Wines, in regard to their alcoholic content, may be divided into two classes — viz., natural -wines, the strength of which has not been increased by the addition of spirit ; and fortified wines, such as those of Spain and Portugal, which absolutely require the addition of a certain amount of spirit to pi-eserve them. Natural wines contain as a minimum 6 per cent., and as a maximum a little over 12 per cent., of absolute alcohol by weight. Even under favourable conditions scarcely more than 15 per cent, of alcohol can be obtained by fermenta- tion. The percentage of alcohol in fortified wines depends, of course, entirely on the operator ; it appears to range usually from 12 to 22 per cent, by weight. The alcohol is returned as ethylic, but there are always traces of the higher homologous alcohols — e.g., propylic, butylic, and amylic. It would be highly desirable to have a number of determinations of the higher alcohols in genuine, and sophisticated wines ; this has never yet been done on any scale. The methods to be followed diifer in no essential way from those already detailed at p. 473, et seq. *The Gf.neral Process of J. Nessler and M. Bartu [Zeit. anal. Chem., 1882, 43). SoiiiL Residue. — The authors evaporate two separate quantities of wine, the one with the addition of a measured quantity of titrated baryta- water, the other without any addition ; the baryta-water mixture is dried at llO^-llo" for eight hours— the addition fixes volatile acid and glycerol. The second jiortion is simply dried at 100° in a current of dry air; the ditierence between the two determinations, correction being made for the baryta, represents glycerol and volatile acid. Effects on Polarised Light. — Neubauer has shown that perfectly fermented wine scarcely polarises light, but if the wine has been imperfectly fermented the plane of polarisation is turned to the left. On the other hand, wines sweetened with potato-sugar contain a considerable proportion of dextro- rotatory non-fermentible substances. The dextro - rotatory substances natural to wine are insoluble in strong alcohol, whilsft the impurities in potato-sugar are mostly soluble in the same. Nessler and Earth have modified Neubauer's process as follows : — The wine is evaporated to one- fifth of its bulk ; the tartaric acid is sepai-ated by precipitation with potassium acetate; 90 per cent, alcohol is next added to complete pre- cipitation; the wliole is filtered, and the filtrate mixed with ether; all optically active substances will now be found in the lower stratum. CIdorine. — The proportion of chlorine in genuine wines lies between "002 and -0028 per cent. , and never exceeds -006 per cent. ; it should not ba estimated in the ash, but directly in the wine. § 278.] VOLATILE ACIDS. 553 § 27S. Volatile Acids* — All wines possess an acid reaction, due to acids, which are conveniently divided into volatile and fixtd. 50 cc. of the wine are acidulated with nitric acid, an excess of standard silver solution added, and then standard thiocyanate solution is run in until a drop of the liquid, when mixed in a plate with ferric sulphate, first shows a pink coloration. Kstimution of Free I'artarxc J cJd. — 100 cc. of the wine are evaporated to a thin syrup, and mixed with alcohol so long as a ])recipitate appears. The cream of tartar, after a few hours standing, is separated. To the filtrate, from 14 to 2 cc. of calcium acetate are added. Wines destitute of free tartaric acid show no turbidity. Any weighable quantity of acid-t^rtrate is, after standing, filtered off. Citric Acid. — In falsified wines citric acid is sometimes met with, 100 cc. of wine are evaporated to 7 cc. , and precipitated with SO per cent, alcohol, and filtered. 'The filtrate is partly neutralised by milk of lime, and the filtrate from this is diluted to the original bulk taken, viz., 100 cc. About 1 cc. of a cold neutral saturated solution of lead acetate is added, and the precipitate (containing phosphoric, sulphuric, tartaric, and part of the malic acid) is collected, decomposed with hydrogen sulphide, and the solution of the free acid rendered alkaline with lime. The calcium jihos- phate is separated by filtration, the filtrate is slightly acidified by acetic acid, tartrate of lime separating. From the filtrate, calcium citrate sepa- rates on prolonged boiling. On drying at 100°, it is weighed — (CsH507)2 Cag + 4H2O. " The acids in wine are estimated by C. Schmidt and C. Hiepe {Zeits. filr anal. Chemie, 21, 534-541) as follows: — 200 cc, of wine, concentrated by evaporation to half, are jjrecipitated by basic lead acetate. The precipitate is filtered oii, suspended in water after being well washed with cold water, and then decomposed by SHo. The solution of the acids thus obtained is filtered from the lead sulphide, concentrated to 50 cc, neutralised with KHO, and still farther concentrated. An excess of a saturated solution of calcium acetate is then added, and the whole allowed to stand for several hours; the precipitate of calcium tartrate is then filtered, washed, ignited, and tlie alkalinity titrated with standard hydrochloric acid. The result is calculated into tartaric acid, a correction of '0286 grm, being added for the solubility of calcium tartrate. The filtrate from the calcium precipitate is again concentrated to 20-30 cc, and 60-90 cc. of 96 per cent, alcohol added. The precipitate consists of calcium malate, succinate, sulphate, and a small quantity of calcium tartrate. It is filtered off, dried at 100°, and weighed. It is then dissolved in the minimum quantity of hot dilute hydrochloric acid, slightly alkalised by potassium carbonate, and the precipitated calcium carbonate removed by filtration. After neutralisation by acetic acid, the filtrate is concentrated to a very small bulk, and j)recipitated hot by barium chloride, which throws down barium succinate and sulphate. The precipi- tate is treated with hot dilate hydrochloric acid, which leaves the barium sulphate undissolved, and which therefore can be filtered off, ignited, and weighed. To the hydrochloric solution of barium succinate, a sufficient quantity of sulphuric acid is added to preci|)itate the barium as sulphate, and from the weight of the latter the amount of succinic acid is calculated, 233 of BaS04 = IIS of C4HCO4. The weights of the sulphuric, succinci, and tartaric acids are calculated as calcium salts, and subtracted from the weight of the lime precipitate, the difference being reckoned as malate, 172 parts of calcium malate =134 malic acid. 554 FOODS : their composition and analysis. [§ 278. Dr. Dupre puts the amount of volatile acid, expressed in terms of acetic acid, as 0-3 to 0-6 per cent, by weight in volume. About one-fourth of the total acidity in white natural wines should be due to volatile acids, and in red and fortified wines the volatile should not amount to more than about one-third of the total acidity. The non-volatile acids appear to be chiefly malic and tartaric (sometimes part of the tartaric being replaced by succinic) ; the former, according to Dupre, predominating in pure natural wines, and largely so in fortified liquors ; whilst in plastered wines it is often present to the total exclusion of tartaric acid. In artificial wines, it is common enough to find a considerable amount of free tartaric acid ; but the mere detection of free tartaric acid is not enough to prove adulteration, since this is . found in small quantity in many natural wines. If, however, with a small amount of free acid there is a preponderance of tartaric acid, then sophistication may be suspected. It has been suggested that such free acid may be recovered from the wine by agitation with ether, but J. Nessler, in a direct experiment, could only recover 3-93 per cent, of the free tartaric acid present when the wine was directly treated with ether; 25 per cent, when the wine was evaporated to a syrup. He recommends the fol- lowing processes : — The wine is agitated with tartar and divided into two parts, to one of which a few drops of concentrated acetate of potassium solution is added, and the mixture carefully observed, noticing whether any tartar crystals form. Errors are avoided by com- paring the one portion to which a few drops of the potassium acetate has been added, with the other portion to which no acetate has been added, the separation of the tartar crystals being a i)roof of free tartaric acid in the wine. Free sulphuric acid may be detected by means of the methods described in the article on Vinegar. The chief volatile acid is acetic ; white German and French wines seldom contain more than -8 grm. per litre of volatile acid reckoned as acetic, and red more than 1-2 per litre. Pted wine containing as much as 1-6 grm. per litre is considered sour and unfit ; it is, therefore, carious to find that the very high-class wines in Table XLIVa may yield as much as 2"15 to 2-42 per litre. The general method of estimation is to take from 10 to 20 cc. of the wine suitably diluted, and titrate with d. n. soda, using tincture of logwood as an indicator, the result being the total acidity. On now evaporating the wine on the water-bath to a syrupy extract, diluting and again titrating, the loss of acidity § 278.] VOLATILE ACIDS. 555 corresponds to the volatile acid, the latter being expressed in terms of acetic acid, the non-volatile as tartaric acid. L. Weif^ert* has shown that by distilling in a vacuum, the whole of the acetic acid can be obtained ; 40 or 50 cc. of the wine are in this way boiled to dryness, water added to the dry residue, and the process thrice repeated. A method of diagnosing and estimating volatile acids has been proposed by E. Duclaux, based upon the more or less regular way in which volatile acids distil each after its own manner, and the estimation of acidity in successive fractions of the distillate. This was described in detail in a former edition, but the method has not found favour among practical chemists and the accuracy of the calculations has been questioned.! Detection and Estimation of Free Suljjhurous Acid and of Aldeliyde-stdplmrous Acid. — Dr. Conrad Schmidt has shown that when a wine is sulphured the sulphurous acid gradually dis- appears, a small part being oxidised to sulphuric acid, the larger part uniting with aldehyde. According to Schmidt free sul- phurous acid is injurious to health, whereas a fair proportion of aldehyde-sulphurous acid can be taken without any injury what- ever. The limit for free sulphurous acid on the Continent is 20 mgrms. per litre. Schmidt has devised a process of estimating free sulphurous and aldehyde-sulphurous acid. A stream of carbon dioxide is led into a 100 cc. flask, and •when the air has been displaced 50 cc. of the wine are added, 5 cc. of sulphuric acid (1:3) and a little starch solution. A decinormal solution of iodine is dropped in until the blue is permanent, and the number of cc. of iodine used is translated into suljjhur dioxide, each cc. of decinormal iodine solution being equal to 2-3 mgms. of SO.,. This determination gives the sulphurous acid existing in a free state. In a 200 cc. flask, 25 cc. of KHO solution and 50 cc. of wine are mixed, and after standing fifteen minutes 10 cc. of sulphuric acid (1:3) and a little starch are added, and the liquid titrated as before. The I'esult is total sulphurous acid, and by subtracting the amount of free sulphurous acid from the total that which has been in combination with aldehyde is known. It is, therefore, only new wine which i's likely to have any free sulphurous acid ; in old wine the sulphurous acid will be in combination. * Zeitsch. fur analyt. Chemie, 1879, 207. t See a paper by Mr. Droop Richmond, Analyst, September and October, 1895. 556 FOODS : their composition and analysis. [§ 278a, § 278a. Estimation of Esters in Wine. — The compound esters in. wine may be divided into volatile and non-volatile. The volatile esters give the bouquet or odour, the fixed esters the taste to wine. The proportion of volatile to fixed esters is very- small in unfortified wines, but the reverse is the case with fortified wines. The total amount of esters is extremely small ; Dr. Dupre gives abou': one part of compound ester in 300 parts of wine as the highest proportion he has yet met with. The esters themselves are, of course, derived from conversion of the alcohols, the ultimate amount depending entirely on the relative proportion of alcohols, acid, and water present, and not being dependent on the nature of the alcohols or acids. If, as some- times happens, an excess of compound ester is added to a wine, decomposition will at once begin, until ultimately the wine will contain no more than it would otherwise have reached in the natural order of things. An estimation of the esters is, there- fore, of the greatest possiV)le importance, as it enables the analyst to judge of the age, character, &c., of the wine. Eerthelot has given the following formula for the calculation of the amount of alcohol present in the compound esters of wine : — y = M7A + 2-28 100 A is the percentage of alcohol by weight in the wine ; a the amount of alcohol equivalent to the total free acid (reckoned as acetic) contained in 1 litre. It hence follows that if the amount of alcohol present as ester, found by experiment, fairly agrees with the calculated amount, etherification is complete, and the wine must be of a certain age ; if the compound esters exceed the proper amount, the probability is that it is an artificial wine; and, lastly, if the amount of esters is below the theoretical standard, either etherification is not complete, on accouut of its youth, or alcohol has been recently added. Since in small quantities of wine the esters cannot be satis- factorily identified. Dr. C. Schmidt's method of expressing tlie esters in terms of decinormal alkali is to be recommended. The wine is first carefully neutralised and 100 cc. are distilled until the distillate measures 90 cc. ; the 90 cc. are made up to 100 cc. by the addition of distilled water; 25 cc. of the 100 are made up to 50 cc. by the addition of neutral absolute alcohol, 25 cc. of d.n. scda added, and the whole allowed to stand for an hour; § 279.] EXTRACT OR SOLID RESIDUE. 557 the alkaline liquid is then titrated, using phenol-phthaleia as an indicator, and a certain loss of alkalinity owing to the saponification of the volatile esters will be found ; this difierence is returned as so much volatile ester in terms of decinormal alkali ; by treating quite similarly the residue in the retort the fixed esters are estimated. The determination of the volatile esters is more important than that of the fixed esters, because the bouquet and properties of a wine are dependent to a considerable extent upon the nature and amount of these esters. Dr. C. Schmidt claims to have synthetised an ester which, added to young wine, gives it the same flavour as an old wine, but no details of his experi- ments have been published. Without a doubt, certain essences used by the wine producers are rich in volatile esters. It is, however, only by an intimate knowledge of the amount of volatile esters which a natural wine may ])0ssess at different periods of time, that it is possible to be certain of the fraudulent addition of esters. III. Estimation and Identification of Matters not Volatile at 100°. § 279. Extract or Solid Residue. — The dry extract in pure natural wines is usually given as from 1-5 to 3 per cent,, some of the Rhine wines of undoubted purity give, however, an extract of nearly 5 per cent. ; the presence of sugar in forti- fied wines may raise the extract to 6-0 or 10 per cent. The solid residue may be taken by simply evaporating 10 cc. to dryness, which can be done rapidly without any decomposition of the solids, by using a large flat platinum dish, and thus spreading the 10 cc. out in a thin layer. This method is, however, somewhat inconvenient, and causes a loss of glycerin ; therefore the indirect process for beer, given at p. 522, may be employed instead, wine extract being considered equal in density to malt extract.* But in wines containing much ash (since the * See also Riegler's method (p. 547). A. Gautier {Annnles d'Hygiene P%iblique, t. xlvii., 118, 1877) has recommended in all cases the evaporation of 5 cc. of wine on a watch-glass, in a vacuum, without the application of artificial heat. This method takes from two to six days, according to the temperature, for completion, so that it is scarcely applicable for technical purposes ; but it is evident that a heat of 30°, whilst greatly expediting the process, would in no way impair its accuracy. 658 FOODS ; their composition and analysis. [§ 279. mineral constituents of the latter seriously affect specific gravity^ containing in a given specific gravity about twice as much substance in solution as a sugar solution of the same gravity), it is necessary to subtract from the percentage of extract thus estimated, the percentage of ash found in the same wine; or if the amount of extract without the ash is required, twice the percentage of ash has to be subtracted from the percentage found. Dupr^ and Thudichum give the following examples : RosENTHALEE, 1859 (£15 Ohm). Specific gravity of de-alcoholised wine, . Percentage of extract (see table, p. 522), . Percentage of ash found, Total solid constituents, , . , . To find total solids minus ash, subtract again . Total solid constituents, .... 3 "782 Sherry, 1865. Specific gravity of de-alcobolised wine, . . . 1017'S Percentage of extract from specific gravity (see table, p. 522), 4-467 Percentage of ash foiind, . . . . . 0'515 3-952 Subtract ash, 0"5)5 Total solid constituents, .... 3-437 H. Hager,* after evaporating off the alcohol, and making up the wine to its original volume by means of water, determines the amount of extract from the following table, which is based on his own experiments, and differs a little from the malt extract table p. 522. The extract and amount of alcohol being known, it is, in cer- tain instances, possible to detect the ivatering of wine, although such a diagnosis can only be made when the analyst is inti- mately acquainted with the kind of wine under examination, and in some cases with the characters of the particular vintage. The Bordeaiix wines, according to Girardin and Pressier, give almost always the same amount of extract, varying only within the limits of 20 to 20-8 grms. the litre; and the projjortion of alcohol also is fairly constant — viz., from -005 to -015, the mean lieing -010 per litre. From these data they calculated the * Chem. Centrhl, 1878, 415. § 279.] ANALYSIS OF WINE. TABLE XLV. 559 Per cent, of Specific Per cent, of Specific Per cent, of Specific Extract. gravity, 15°-0 Extract. gravity, 15^-0 Extract. gravity, l&o-O 0-50 1-0022 5-25 1-0239 10-00 1-0461 075 r0034 5-50 1-0251 10-25 1 0473 100 1U046 5-75 1-0263 10-50 1-0485 1-25 10057 6 00 1-0274 10-75 10496 1-50 1-006S 6-25 1-0286 11-00 1-0508 1-75 1 -0079 6-50 1-0298 11-25 1-0520 2 00 1-0091 6-75 1-0309 11-50 1-0532 2-25 10102 700 10321 11-75 10544 2-50 1-0114 7-25 1-0332 1200 1-0555 2-75 1-0125 7-50 10343 12-25 1-0567 3 00 1-0137 7-75 1-0355 12-50 1-0579 3-25 1-0148 8-00 1-0367 12-75 1-0591 3-50 1-0160 8-25 10378 13-00 1-0603 3-75 10171 8-50 1-0390 13-25 10614 4-00 1-01S3 8-75 10402 13 50 10626 4-25 1-0194 9-00 10414 13-75 1-0638 4-50 1-0205 9-25 10426 14 00 1-0651 4-75 1-0216 9-50 1-0437 14-25 1-0663 5-00 1-0228 9-75 1 -0449 or 72500 [N.B. — The specific sn'avity increases or diminishes "00024 for each degree.} amount of genuine wine present in any sami^les. Thus, suppos- ing the extract in a Bordeaux wine to be 14-5, then 1000 X 14-5 20-9 i.e., the litre contains 725' cc. of wine, the rest being alcohol and water. To know the quantity of alcohol added, it is necessary ta ascertain how much the 72-5 parts of wine contain of absolute alcohol ; 100 : 10 : : 72-50 : x X = 7-25. If the absolute alcohol is found, for example, to be 0-11, then^ subtracting 7-25 from 11, it is supposed that 3-75 of alcohol has been added. That this process, as applied to the Bordeaux wines, is in the main correct, is supported by the fact that the Bovien wine-mer- chants have frequently paid duty on the excess of alcohol, itc, which Girardin and Pressier found in their wines.* * In Paris a commercial standard has been arrived at, based on the analysis, of 6,000 samples, and it is laid down that " the amount of added water in all wine which is not sold as being of any special brand, shall be calculated on the basis of 12 per cent, of alcohol by volume, and 24 grms. of dry extract per litre." 560 FOODS : TiiEin composition and analysis. [§ 279. In the Municipal Laboratory, Paris, the chemists determine the watering and fortifying of wine by a calculation of the relationship of what they call "the reduced extract" to alcohol. The "reduced extract" is the total extract diminished by the number of grammes less 1 of the potassic sulphate and reducing sugar; thus, if the total extract should be 29-7, the potassic sulphate 3-1, and the sugar 4-5; 2-1 + 3-5 = 5'6 ; and subtract- ing 5 '6 from tlie total extract gives the reduced extract as 2-i'l. The weight of the alcohol for red wines they consider should not exceed four and a-half times tlie extract, and when this relation is exceeded the wine lias been fortified. To deter- mine this relation, the alcohol by weight is divided by the reduced extract. For white wines the relation between alcohol and reduced extract is fixed at 6-5. Wines watered down and then the alcoholic strength brought up by the addition of alcohol are detected in the French Laboratory as follows : — In all normal wines the sum of the alcohol per cent, by volume added to the total acidity per litre (alcohol-acid number), calcu- lated as sulphuric acid, is seldom below 12"5. Water lowers this number, alcohol increases it. First, then, the relation of the alcohol to the reduced extract is obtained; if it exceed 4-5, then by calculation on the standard of 4-5, the amount of alcohol that the natui-al wine may be supposed to have had originally is obtained; and the difierence between this and the amount found represents the alcohol added. Next, the alcohol-acid number is obtained, and if this is below 12-5 the presumption is that the wine has been watered. An example will make this clear. A red wine gave the following : — Dry extract per litre, 14'2 Acidity per litre, 3-100 Alcohol per cent, (volume), 16'0 The relation by weight of alcohol-extract, . . = 9"01 The alcohol-acid number, =19-1 I^ow calculating on a standard of . . . . 4 5 Natural alcohol in the wine, . . . 14'2 x 45 = G3 9 Iveducing this by dividing by 0'8, the volume of alcohol is equal to . . . . . . 7 '99 Alcohol in excess, 16-7-99= 801 Alcoliol acid fii^uie, 7-99 -t- 3-1 = 11-09 Hence, the alcohol-extract number being superior to 4-5 and the corrected Acid-alcohol number being below 125, there is a presumption of both watering and fortifying. This calculation being the results of prolonged experience with regard to ordinary wines is a fairly safe guide to the § 2S0.] ESTIMATION OF SUCCINIC ACID AND GLYCERIN. 561 analyst ; but with regard to certain exceptional wines caution must be exercised in the interpretation of results.* § 280. Estimation of Succinic Acid and Glycerin. — Half a litre to a litre of wine is decolorised with animal charcoal, filtered, and the chai'coal well washed with water; the filtrate and wash- ings are then evaporated down in the water-bath, and the drying finished in a vacuum. The residue, when dry, is treated with a mixture of 1 part of strong alcohol and 2^ parts of rectified ether. The latter is driven oft' by floating the dish in warm water, and the wLole evaporated again on a water-bath. The residue is now neutralised with lime-water, which combines with the succinic acid, and forms succinate of calcium. The glycerin is dissolved out by alcohol and ether, and weighed either directly or by loss. The succinate of calcium remaining behind is impure, and should be well washed with spirit before weighing. Every 100 parts of calcic succinate equals 75-64 of succinic acid (HoO^H^O^) ; and since Pasteur has shown that 112-8 parts of grape-sugar (107 of cane) yield about 3'6 of glycerin and O'G part of succinic acid, it follows that in a natural wine the glycerin would amount to about one-fourteenth part of the alcohol present, f It has, indeed, hitherto been generally accepted that for every 100 parts of alcohol there shoukl be not less than 7 nor more than l-l of glycerin, and it has been held that deviations from this standard mean addition of alcohol or glycerin ; but the highest quality of the Rhine wines (Table XLlVa) vary between 16 and 31 parts of glycerin to 100 of alcohol, and, therefore, the views hitherto held demand modification. SchmidtJ evaporates a known bulk of the wine with hydrated calcium oxide, extracts the residue with 96 per cent, alcohol, evaporates the clear filtrate, then dissolves this last residue in 15 cc. absolute alcohol, precipi- tates with 25 cc. of ether, filters and evaporates the alcohol-ether solution, and, after one and a-half hours' drying, weighs. Stierlin § evaporates the liquid without any addition to one-fifth or one-sixth of its volume, extracts with hot absolute alcohol, and uses this alcoholic extract for the estimation of the sugar, non-volatile acids, bitter matters, * E. Egger detects the watering of wine by the presence of nitrates, the grape being stated to be absolutely destitute of nitrates ; white wines are evaporated to a syrup and absolute alcohol added ; so long as it produces a cloud the mixture is filtered, decolorised, and tested with dipheuylamine and sulphuric acid. Ked wines are precipitated with lead acetate and then white magnesium sulphate before evaporation [Chtm. Centr., 1885, 71, 72). t According to E. Borgmanu {Zeilsch. fur. anal. Chemie, xxii., 58-60) the ratio of glycerin to alcohol in pure wines is never less than 7 '3 : 100. Analyses of white wines by R. Fresenius and E. Borgmann give the follow- ing ratios of glycerin and alcohol : — Alcohol. Glycerin. Maximum, lUO : 14-4: Minimum, 100 : 7*3 Mean, 100 : 10*4 (Zeit. f. anal. Chemie, xxiii., 48). 1 Op. cit. % Stierlin, "Das Bier," &c. Bern, 1879. 37 562 FOODS : their composition and analysis, [§ 281. alkaloids, and glycerin. For the estimation of the last, a measured portion of the alcoholic extract is freed from alcohol by evaporation, and then eva- porated down to dryness with slight excess of caustic lime. The glycerin is extracted either with alcohol and ether (2 : 3), or with alcohol and chloro- form. (See also the process for extracting glycerin from beer, p 5.31.) Raynaud has pointed out that although the processes in use for the esti- mation of glycerin are fairly exact, yet with plastered wines too high results are obtained ; for if there is any considerable amount of sulphate of potash, it is decomposed by lime, and hydrate of potash is formed, which is dissolved by glycerin in the presence of alcohol, and is weighed with it. He therefore recommends the following process : — The liquid operated upon is evaporated to about one-fifth of its volume, and the pobash precipitated by hydrofluo- silicic acid and filtered. The filtrate is made weakly alkaline by the addition of hydrate of baryta ; sand is also added, and the mass is evaporated to dry- ness in a vacuum ; the dry residue is then extracted with a very large volume of absolute alcohol and ether, as much as 300 cc. for 250 cc. of wine being recommended. With the improved processes of extraction which we now possess, however, this is quite unnecessary, and 50 to 100 cc. in a Soxhlet's apparatus (see p. 67) will have quite the same effect as a much larger quantity. On the evaporation of the alcohol and ether, the glycerin is allowed to stand for twenty-four hours in a vacuum over phosphoric anhydride ; finally, it is put into a tube, a perfect vacuum formed, and distilled into the cool part of the tube by a temperature of 180°. Probably the best method of estimating glycerin is to separate it from most volatile substances by distillation in a vacuum, and then to oxidise it into oxalic acid, as described under the article Butter (p. 371). A. Partheil* effects this in the following manner: — 50 cu. of the liquid to be examined, first neutralised by adding a little calcium carbonate, are evaporated down to 15 cc. and introduced into a small retort. This retort is enclosed in an air bath, the bottom of the bath being made of sheet iron, the sides and top of asbestos card. The neck is connected with a globular receiver, the second opening of the receiver being joined to an inverted condenser, and then to a pump. The receiver is also kept cool. The liquid is first distilled almost to dryness, at ordinary pressure, at a temperature of 12IJ°. It is then cooled to about 60°, and the pressure reduced by means of the pump, the temperature raised to 180°, and the distillation continued for one and a-half hours; the pressure at the end of that time is released, the retort cooled, 10 cc. of water added, and distillation again proceeded with at the ordinary pressure at a temperature in the bath of 120°. The distillate is diluted to about 200 cc, 8 to 10 grms. of caustic soda dissolved in it, and 5 per cent, potassic permanganate added until the colour remains a decided blue-black. The whole is heated for an hour, decolorised with SOo, 20 cc. of acetic acid added, the SO2 driven off by heat, and the oxalic acid precipitated by calcium chloride. § 281. Estimation of Tartaric Acid and Bltartrate of Potash. — This is best estimated by the method suggested by Berthelot : — 20 cc. of wine are mixed with 100 cc. of equal volumes of alcohol and ether in a well-stoppered flask. The same process is employed to another 20 cc, but with the addition of potash in sufficient quantity to neutralise about one-fifth of the free acid present. Both bottles are allowed to stand two or three days, and at the * Arch. Pliarm., 1895, ccxxxiii., 391. § 281.] ALBUMINOID SUBSTANCES. 56-3 end of the time, owing to the insolubility of bitartrate of potash in strong alcohol, there will be a deposit of that salt in both bottles. ' The first will represent the bitartrate of potash present as such ; the second, the whole of the tartaric acid which the wine contains. There is, however, always a small quantity of bitartrate in solution, about '004 grm., equalling -28 d. n. soda, and this amount must be added to that found. The precipitates from both bottles are collected on separate filters, washed with the alcohol-ether mixture, dissolved in water, and titrated with soda solution. Direct Estimation of Malic Acid. — 100 cc. of wine are precipi- tated with lime-water, added only in slight excess ; the filtrate is evaporated down to one-half, and absolute alcohol added in excess ; the resulting precipitate, consisting of malate and sul- phate of lime, is then collected on a filter, washed, and weighed. If, now, the sulphate of lime in this sample be estimated by solution in water, and precipitation of the sulphuric acid by baric chloride, kc, and the amount subtracted from the total weight of the precipitate, the remainder equals malate of lime. The Esthnation of Sugar in Wine is carried out on the prin- ciples described at p. 137, et seq* Albuminoid Substances. — The albuminoid substances in wine may be estimated by Mr. Wanklyn's well-known ammonia pro- cess : — 5 cc. of the wine are put in a half-litre flask, and made up with water to 500 cc. : yl^ (i.e., 5 cc.) of this is distilled with a little water and pure carbonate of soda (ammonia free), and the ammonia in the distillate estimated by the colorimetric process known as Nesslerising. An alkaline solution of permanganate of potash is then added, and the operation repeated — the ammonia coming over now being the result of the breaking-up of albumi- noid bodies. It would appear that in ivltite wines, the albuminous matters are very small in amount ; while in red and most young wines, there is an excess of albuminous matters, which decreases * A special process is in use in the Paris Laboratory for the detection of fictitious claret. The Hquid is known as piquette of raisins and dried fruits, and it is often added as an adulterant to the genuine article. 300 cc. of the wine are fermented fully with yeast at 29° '5. The liquid is then filtered, and placed in a dialysing apparatus, in which the outer water is constantly renewed automatically. The water is examined from time to time by the polariscope, and, when the polarisation is constant, the dialysis is stopped. The liquid is neutralised by boiling M'ith chalk, and evaporated to dryness on the water-bath. The residue is treated with 50 cc. of absolute alcohol, and twice washed with 25 cc. of the same. The alcoholic extract is next decolorised by charcoal, evaporated to dryness, and the residue taken up with 30 cc. of water and exannned by the polariscope. True claret gives no rotation, or is only very slightly dextrogyrate, while wines mixed with piquette of fruit or glucose, are respectively strongly levo- or dextro- gyrate. — Dr. Muter in Analijut, October, 1885. 564 foods: their composition and analysis. [§§ 281a, 282. with age ; hence, in experienced hands, a determination of this kind may help to distinguish between old and new. Thudichum and Dupre found in certain wines the following amounts of ammonia : — Ammonia free. Ammonia albd. Per cent. Per cent. Ingelheimer, red, . 0-0051 0-3730 Port, 1851, . 0046 0-0888 Sherry, 30 years in bottle, . 0-01)73 0-1807 Madeira, .... . 0-0021 0-1581 Merstemer, . 0-0021 0-3550 Natural Port, . . 0-0019 00527 Port, 1865, . 00012 01760 § 281a. Astringent Matters. — Schmidt estimates the astringent and colouring-matters of wine by oxidising with potassic per- manganate before and after treatment with animal charcoal. A. Girard {Compt. Rend., 95, 185-187) employs sheep-gut for the estimation of tannin in wine. The gut is well washed and cleaned ; it is then treated with alkalies, and bleached by the iiction of potassic permanganate and sulphurous acid. It is then twisted into cords, and again bleached by sulphurous anhydride. From 3 to 5 grms. of the gut-cords (the water in which has been pi-eviously determined) are soaked in water for four or five hour.s, and, after being untwisted, are added to 100 cc. of the wine. After from twenty-four to forty-eight hours, the whole of the tannin is absorbed, and is ex])ressed by the gain in weight of the gut after being washed and dried at 100°. § 282. Estimation of the Colouring -matter of Wine. — The colouring -matter of red wines has been termed mnolin, or oenocyanin, and has also received other names. Glenard has assigned to it the formula C^gHj^Og ; but it is doubtful whether it has ever yet been separated in a state of absolute purity. The process used by Glenard was — precipitation with lead acetate, exhaustion of the washed, dried, and powdered precipitate, first, with anhydrous ether saturated with HCl, then with pure ether; and, lastly, extraction with alcohol, from which the oenolin was obtained by evajjoration as a bluish-black powder insoluble in ether, almost insoluble in pure water, but more readily dissolved in acidulated water, acidulated alcohol dissolving it easily. The blue colour is turned red by acid. Q^nolin, according to Vaserine,* may be separated from wine by mixing the latter with lime to the consistency of a paste, which is drained on a funnel. The residue, containing the colouring-matter, is mixed with alcohol of 95 per cent., and treated with sufficient sulphuric acid to neutralise the lime and decompose the compound of lime with the colouring -matter. The solution is filtered from calcium sulphate, and on evaporation leaves oenolin as a black * Bull. Soc. Chim. [2], xxix., 109, 110. § 282.] ESTIMATION OF THE COLOURING-MATTER OF WINE. 565 powder.* Solutions of oenolin. show, when examined by the spectroscope, certain bands. The colour of white wines is due to oxidised tannin ; it takes long to develop; hence the manufacturer not unfrequently adds a little caramel. Shoukl this be the only addition, it would be injudicious to consider the wine adulterated. The ai-tificial colouring of wines by elder-berry, logwood, cochineal, aniline, &c.,f is said to exist, at all events, on the Continent ; and it is u fact that a few home-made, low-priced wines, almost entirely fictitious, are passed oft' by the aid of the same or similar substances ; but with regard to the ordinary foreign wines in English commerce, there is no reliable evidence whatever that any adulteration of this kind has been practised. Nevertheless, it is absolutely necessary to be aquainted with the best and most recent methods for the discovery of such frauds. The substances actually found to be used fraudulently as artificial colouring are — (1.) Bordeaux verdissant, which is a compound of methylene blue, diphenylamine orange, and the acid-sulpho-derivative of fuchsine ; (2.) a mixture of amido- benzene, methyl- violet, and the acid-sulpho-derivative of fuchsine. In the Paris Laboratory J they use three preliminary tests, and consider wines genuine as to colour if they respond to these tests. On the other hand, should the results be imfavourable, the analyst possesses a valuable guide to the class of colouring- matters present. The three methods are as follows : — (1.) Sticks of chalk are steeped in a 10 per cent, solution of egg-albvimen, and dried first in the air and then at 100°. The wine is tested by allowing two drops to fall on a siirface of the chalk from which the excess of albumen has been removed by scraping ; genuine wine gives a gray colour, and young and highly-coloured wine may give a somewhat bluish tint, but there should be no trace of green, violet, or rose. (2.) The wine is alkalised by baryta-water until it is of a * According to L. M. Krohn, the red colouring-matter of wine may be obtained by electrolysis as a deposit on the positive pole. The colouring- matters used to adulterate wines do not give this deposit (Journ. Fharm., (5), ix., 298-300), f Soubeiran says — " At Fismes, in the neighbourhood of Rheims, there has been manufactured for more than a century (since 1741) a colouring agent composed of elder-berries, alum, and water, in different proportions, the pro- longed use of which can only have injurious consequences on the health, on account of the alum. Unfortunately, the production of this colouring agent {teinte) was encouraged by a royal decree of 1781. . . . Recent analyses have shown that this liqueur de Fiiimes contains from 20'8 to 57'5G grms. per litre of sdam^—Dict. des Falsifications. Paris, 1874. + " Report on processes in use at the Municipal Laboratory of the city of Paris," Ey Dr. Muter. Analyst, 1885. 566 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 282. greenish line; it is then shaken np with acetic ether or amy lie alcohol. If the wine is pure there is no colour in the upper layer, with or without the addition of acetic acid. On the other hand, coal-tar colours of a basic nature colour the solvent, and give indications suggestive of amido-benzene, fuchsine, safranine, chrysoidine, chrysaniline, mauveine, methyl-violet, and Beibrich red. (3.) 10 cc. of the wine are alkalised until the wine becomes of a green colour by the addition of 5 per cent, potassium-hydrate solution. To this are added 2 cc. of a solution of mercurous acetate ; the whole, after shaking, is filtered. With pure wine the filtrate is colourless, both in itself and after acidulating with hydrochloric acid, while coal-tar colours of an acid nature tint the filtrate red or yellow. If either of the two last tests responds, the wine is further examined as follows : — A. — TnfRE IS EVIDENCE FROM (2.) OF AN Aniline Basic Colour. The amylic alcohol or acetic ether solvent is sefiarated and divided into two parts, W. and S.- — the one, W., is evaporated down with threads of wool, the other, S., on threads of silk. The wool and silk are both coloured red, hydrochloric acid dis- colours it to a dirty brown, but water restores the red. Rosanil'ine. The silk is coloured red, hut not the wool; the reaction with hydro- chloric acid does not take place. Safranine. Violet, and when treated with HCl, becomes bluish-green, changing to yellow, but water in excess restores the violet. Soluble Aniline Violets. Indigo-blue, when treated with HCl, but on dilution with water, reddish-brown. 31auvaniline. Threads scarcely affected with HCl, but decolorised by boiling with pow- dered Zinc, the colour returning on exposure to air. Chrysotoluidine. Straw-yellow threads, becoming poppy-red with strong sulphuric acid. Amidonitrobenzol. B. — There is evidence from (3.) OF A Tar Dye of an Acid Nature. The wine is strongly acidulated with hydrochloric acid, and shaken up with acetic ether (or amylic alco- hol). The wine is saturated with a slight excess of ammonia, and shaken up with the same solvent. The ethereal fluids are mixed, evaporated to dryness, and tested with a drop of strong sulphuric acid. Parma, violet colour. lioccelline. Maroon. Foundation Red. Blue. Bordeaux R. and B. Scarlet. Ponceau J?. Red. „ B. Green to violet. The Beibrich Red. Fuchsine red. Tropcollne 000. Orange yellow, or on dilution with water a transient poppy-red. Tropeoline O., or Chrysoinc. Brownish-yellow. Helianthine. Yellow. Eosine B., or J. J. Safranine. Ethykosine. § 282.] ESTIMATION OF THE COLOURING-MATTER OF WINE. 567 Orange-yellow threads, becoming scarlet with sulphuric acid. Chrysoidine. Threads of a reddish colour, decol- orised instantly by a few drops of sodium bisulphite. Dyes analogous or allied, to Rosaniline. Dr. Dupre, taking advantage of the fact that the colouring- matter of wine* only dialyses to a minute extent, and that the colouring-matters of Brazil, logwood, and cochineal, readily dialyse, separates the latter colouring-matters from the wine by dialysis. The same chemist has suggested a still more convenient and practical process — viz., the staining of a jelly. The jelly is made by dissolving 5 grms. of gelatin in 100 cc. of warm water, and pouring the solution into a square flat mould made of paper. From this cake cubes abovit | inch square are cut with a sharp wet knife, and are immersed in the wine, taken out after the lapse of from twenty-four to forty-eight hours, washed slightly, and sections cut, in order to see how far the colouring principles have penetrated. If the wine is pure, the colour will be confined almost entii-ely to the edges of the slice, or will not have pene- trated beyond Jg- to ^ inch ; most other colouring-matters rapidly permeate and colour the jelly. (1.) Colouring-matters penetrating slowly into tliejclhj: — Colouring-matter of pure wine. „ ,, Ehatany root. (2.) Colouring-matters penetrating rapidly into the jelly : — Rosaniline. Cochineal. Logwood. Brazil-wood. ludigo. Litmus. Bed Cabbage. Beetroot. Malva sylvestris. Althea officiiialis. The jelly may be examined spectroscopically, good results being obtained in the case of rosaniline, red cabbage, and beet- root ; and may be also tested with reagents — e.g., dilute ammonia dissolves much colour from tlie slice, if the colour should be derived from logwood or cochineal ; on the other hand, the • The colouring-matter of Rhatany root has the same property. 5G8 FOODS : their composition and analysis. [§ 282. ammonia remains colourless in the case of rosaniline, red cabbage, and beetroot. A simple method for the detection of certain colouring-matters is that of Lammatine: — Shake 100 parts of the wine with 100 of coarsely-powdered peroxide of manganese, and then pass through a double filter; if pure, a colourless filtrate will result. The process is said to answer well in the case of logwood and cochineal, but to fail with aniline. A general method, applicable to fuchsine and other colouring- matters, is based upon the fact that a great many of the colour- ing-matters which may be used for purposes of wine adulteration (such as caramels, ammoniacal cochineal, sulphindigotic acid, logwood, and the lichen reds), are precipitated by acetate of lead; whilst fuchsine, if present in considerable quantity, is only partially thrown down. Those which are not precipi- tated may be separated by agitating the filtrate with amyl alcohol. The lead precipitate may be treated by dissolving out cochi- neal, sulphindigotic acid, and fuchsine, by a solution of potassic carbonate [2 : 100]. From this liquid the fuchsine is separated by neutralisation with acetic acid and agitation with amyl alcohol, the rose-coloured liquid obtained being identified as a solution of fuchsine by the spectroscope. On now acidifying the liquid with sulphuric acid, the carminamic acid from the cochineal is removed by means of amyl alcohol, and identified by its three bands — viz., one between D and E in the red, the second in the green, and the third in the blue. The indigo remaining may be detected by the blue colour, and absorption- band between C and D. The original lead-precipitate, insoluble in potassic carbonate, is treated with a two per cent, solution of potassium sulphide, which dissolves the colouring-matter of logwood and that of the wine itself. Logwood may, however, be tested for directly in the wine by the addition of calcium carbonate and two or thi-ee drops of lime-water. In the case of a natural wine, the filtered liquid is almost colourless, but is of a fine red colour if logwood is present. Lastly, the lichen red may be obtained by washing the insoluble portion left after treatment with potassium sulphide, and dissolving it in alcohol, when a red colour and a definite absorption-band reveal its presence. For the detection of fuchsine simply. Bouillon {Comptes rendus, Ixxxiii. 858, 859) recommends half a litre of the wine to be evaporated down to 120 cc, with the addition of 20 grms. of barium hydrate. It is then filtered, and the filtrate shaken up ■with ether; the ether is separated, a drop of acetic acid, a little water, and a small piece of white silk, are added, and (if an appreciable amount of fuclisine is present) the silk assumes a pink colour immediately; if not, the liquid must be concentrated nearly to dryness. F. Konig* has a process for detecting fuchsine: 50 cc. of the wine are treated with ammonia in slight excess, and boiled with a little pure wool [-5 grm.], until all the alcohol and ammonia are evaporated. The wool is washed and directly moistened with strong potash, and heated until it dissolves into a more or less brown fluid. After cooling, to this is added half its volume of pure alcohol, and then an equal volume of ether; it is strongly shaken. The smallest trace of fuchsine is taken up by the ether, and is coloured red by acetic acid. 0*4 mgrm. fuchsine in a litre of wine is said by this means to be discovered readily. The process destroys the natural colour of the wine. § 283. Vogel and others have studied the detection of the colouring-matters of wine by means of the spectroscope. The following curves are examples of various colouring-matters (see lig. 66):— No. 37. Wine colouring - matter, (I.) pure, (II.) diluted. No. 38. Wine coloui'ing - mattei', with ammonia. No. 39. (I.) Mallow colouring - matter concentrated, (II.) elderbei-ry concentrated. No. 40. Acid cherry, (b) Acid cherry with the addition of tannin. No. 41. Mallow colouring-matter with the addition of alum. No. 42. Indigo solution. ^ With carefully made comparison solutions, there can be little doubt that the spectroscopical method of identifying colouring-matters will be found of great value. Lastly, M. A. Gautierf has proposed a method aiming at a systematic detection of every probable colouring-matter likely to be added to wine. How far the whole, or any portion, of this elaborate system will be followed and confirmed by chemists remains to be seen. The following abstract of M. Gautier's paper is taken from the Analyst, i., 1877: — * Ber. der deutsch. Chem. Gesdlsch., Berlin, xiii. 2263. t /. Pharm. Chim. [4], xxv. 8-12 and 102-106. 570 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 284. M. Gautier's Process for Detecting Colouring-Matters IN Wine. § 284. Preliminary Preparation of the Sample. — The sample is mixed with y"^ its volume of white-of-egg previously diluted with \h times its bulk of water, well shaken, and, after standing for half an hour, filtered. If the wine is very poor in (annates, a few drops of a fresli aqueous solution of tannin should be added before agitating with albumen. The filtrate is treated with dilute sodium bicarbonate until its reaction is very feebly acid. All the reactions of Table B. must be made on this liquid, except those for indigo, which are executed upon the albuminous precipitate. Table B. — systematic process to be followed for the detec- tion OF THE NATURE OF FOREIGN COLOURING-MATTERS ADDED TO WINES. A. Having placed aside the filtrate from the albuminous pre- cipitate, the precipitate is washed until the washings are almost colourless. Two cases may present themselves : — (a.) The precipitate after washing remains wine-coloured, lilac, or maroon ; loine, natural, or adulterated tvith the greater part of the substances usually employed. Pass on to C. (6. ) The precipitate is of a very deep wine-colour, violet-blue, or bluish; toinesfrom, tlie deepest coloured grapes, or wines coloured ■with indigo. Proceed to B. B. The precipitate is washed with water, then with alcohol of 25 p.c, a part is then removed and boiled with alcohol of 85 p.c. (a.) The filtrate is rose, or wine-coloured. A portion of the precipitate is removed from the filter, suspended in water, and carefully saturated with dilute })otassium carbonate. The colour changes to brown or blackish-brown; natural loiries, or adulterated tvith substances other than indigo. Pass to C. (h.) The filtrate is bhie. A portion of the precipitate sus- pended in water and treated with dilute potassium carbonate affords a deep blue liquid, which changes to yellow by an excess of the reagent. Various preparations of indigo. Indigo. C. 2 cc. of wine are treated with 6 to 8 cc. of a -o^ „ solution of sodium carbonate, which must be added in slight excess (1 cc.) after the change of colour. («.) The liquid becomes lilac, or violet, sometimes the liquid becomes only winey, or dashed with violet. Brazil-wood, cochineal, Portugal berries, fuchsine, . . . wines of certain sorts, fresh beetroot, logwood, both elders, whortleberries (myrtille), Portugal ■berries. Pass to D. § 284.] COLOURING-MATTERS MIXED WITH WINE. 571 (b.) The liquid becomes bluish-green, sometimes with a faint lilac tint, zvine, hollyhock, privet,'* whortleberries, logv)ood, Portugal berries, fuchsine. Pass to M. (c.) The liquid becomes greenish-yellow without any blue or violet, beetroot (old or fermented decoction), whortleberries, certain rare varieties ofioine. Pass to L. D. The liquid 0. («.) is heated to boiling. (a.) The liquid remains wine-violet, rose, or wine-lilac, or becomes a brighter lilac ; logwood, Brazil-wood, cochineal, certain varieties of icine. Pass to E. (b.) The colour disappears, or changes to a yellow, or maroon, or reddish tint, wine, fuchsine, both elders; whortleberries, Portugal berries, fresh beetroot. Pass to F. E. Treat 4 cc. of the wine with 2 cc. of each of a 10 per cent, solution of alum, and a 10 per cent, solution of crystallised sodium carbonate. Filter. (a.) Clear yellowish-green lake (which may be bluish from mix- tures of wines containing maroon), filtrate colourless, becoming very slightly yellow on warming ; its own volume of aluminium acetate at 2° B. almost wholly decolorises it. On acidification with acetic acid, after treatment with its own volume of barium hydrate (saturated solution), the wine becomes clear greenish- yellow or maroon, j^ure or mixed wines. See Table A. (6.) Greenish-blue lake, or dirty yellowish-green, according to the varieties present, sometimes very slightly winey. Filtrate bright-rose, gradually decolorised on warming, though retaining a tinge of lilac ; not decolorised by lime-water in the cold. Cochineal. (c.) Winey- violet lake, which darkens on exposure to the air. Filtrate bottle-green, or grey faintly red (if much logwood is present). The filtrate becomes green on warming. Logwood. {d.) Lilac, or maroon-lilac lake. Filtrate greyish with tint of maroon. On boiling, this filtrate becomes fine old- wine coloured. Brazil-wood. F. Treat 4 cc. of the wine with alum and sodium carbonate (as explained at E.), add to the mixture two or three drops of very dilute sodium carbonate, and filter. (a.) The filtrate is lilac or winey, Portugal berries, fresh beetroot. Pass to G. * The colouring-matter of privet berries is stated to be used iu Saxony for colouring wine ; it gives an absorption band at D., and a faint absorption at F. This colouring-matter is extracted by amyl alcohol. Tartaric acid heightens the colour, shutting out all the blue. Alum colours it beautifully blue and broadens the absorption at D., while the absorption of the blue and green is diminished. Tartaric acid annihilates the blue colour, and gives a colour similar to wine ; careful neutralisation with ammonia restores the blue colour and the band at D ( Vogel). 572 foods: their composition* and analysis. [§ 284. (b.) The filtrate is bottle-green, or reddish-green, wine, fuchsine, black elder, whortleberries, beetroot. Pass to H. G. Treat 2 cc. of the wine with subacetate of lead solution, of density 15° B. Shake. Filter. (a.) The filtrate is rose, which persists even wlien made slightly alkaline ; it slowly disappears on boiling. Lime-water destroys the rose colour. Portugal Berries. (6.) The filtrate is yellowish, or brownish-red. Fresh Beet- root. H. The alum-lake obtained from F. (5.) is — («.) Deep blue. On treating the clarified wine with a few drops of aluminium acetate solution, it becomes a decided violet, or wine violet. Both elders. Pass to I. (6.) Bluish-green, green, or faintly rose-tinted, loine, whortle- berries, beetroot, fuchsine. Pass to J. I. After the test H. (a.) treat a fresh quantity of 2 cc. with 1-5 to 2 cc. (according to its acidity and the depth of its colour) of an 8 per cent, solution of sodium bicarbonate charged with carbonic acid. {a.) The liquid remains lilac for a moment, then changes to greenish-grey blue. Another specimen treated with sodium carbonate (according to C), and heated to boiling, becomes dark greenish -grey. Black Elder. (6.) The liquid retains a lilac tint, or becomes grey with mixture of maroon, or dirty lilac. Another specimen treated with sodium carbonate (as at C.) tends to discolour on heating, the green being replaced by red. Dwarf Elder. J. Treat 5 cc. of the clarified wine with a slight excess of ammonia, heat to boiling, and after cooling shake with 10 cc. of ether, decant and evaporate the ether, and treat the residue left on evaporation with acetic acid. {a.) The liquid becomes red. Fuchsine. (6.) The liquid does not become red, wine, lohortleberries, fresh beetroot. Pass to K. K. Another specimen is treated according to C. with sodium carbonate. (a.) The colour darkens or becomes red on heating, lohortle- berries, fresh beetroot. Pass to L. (b.) The greenish or bluish-green liquid, possibly having a winey tinge, has a tendency to discolour on heating. Natural wine. L. Treated with sodium bicarbonate according to the rules given at I. (a.) The liquid is deep grey, slightly greenish, green, some- times green with very slight lilac tint. § 284.] COLOURING-MATTERS MIXED WITH ^YI^•E. 573 The clarified wine, treated with an equal volume of saturated baryta water, and filtered after standing for fifteen minutes, gives a dirt}' yellow, or slightly greenish filtrate. With an equal volume of aluminium acetate of 2" B. it gives a lilac wine-coloured filtrate. With a few drops of aluminate of potash no change of colour. With sodium carbonate, enijiloyed as at C, the liquid tends to lose its colour on heating. With barium peroxide, used accord- ing to Table A, column P, the liquid is faintly rose-tinted, with or without an orange-coloured deposit on the barium peroxide. Natural Wine. With the general characters above indicated, if with baryta water it affords a madeira-coloured filtrate, changing to buff on acidulation with acetic acid ; if with borax it becomes deep-green with a bluish cast ; if with alum and sodium carbonate (as at E) a precipitate falls of a deep bottle-green, with bluish tinge, and if with aluminium acetate it remains rose-coloured with no change to violet-blue. Teinturier. (6.) The liquid is reddish-yellow or brown-lilac. By ti-eat- ment with acetate of alumina the filtrate is clear lilac. With a few drops of aluminate of potasli the colour becomes that of the skin of an onion, and with a larger quantity of the reagent the colour is green, tinged with maroon. With sodium carbonate (employed as at C.) the fluid passes to yellowish or greyish-yellow, with tinge of red. With barium peroxide, flesh-coloured liquid with considerable orange-coloured deposit in contact with the peroxide. Beetkoo'p, fermented or not. (c.) The liquid is j'ellowish-grey, with tinge of green or red. With baryta water the filtrate is yellowish olive-green. With aluminium acetate the filtrate is bluish-violet, or violet-lilac. With aluminate of potash, fresh rose, becoming yellowish-gi-een, with an excess of reagent. With sodium carbonate (as at C.) the fluid becomes deep grey on heating. With barium peroxide the fluid is bleached, or remains but very slightly roseate, with a trace of orange deposit in contact with the peroxide. Whortle- berries. M. The mixture of wine and alkaline carbonate C. (6.) is heated to boiling. (a.) Tlie mixture becomes lilac-violet, or violet. Logwood. (6.) The mixture tends to become decolorised, or changes to yellowish-green, or dark green, or maroon green, natural wines, ivhortleberries, both elders, privet, Portugal berries, fuchsine. Pass toK N. Treat the wine with alum and sodium carbonate, as directed at E., and filter. 574 foods: their composition and analysis. [§ 284. {a.) The coloTir of the filtrate is lilac. Portugal berries. (b.) The filtrate changes to bottle-green, or reddish-green. Natural wines, lohortleberries, hollyliocJc, privet, both elders, fuchsine. Pass to O. O. Treat 2 cc. of the clarified wine with 3 or 4 cc. of a saturated solution of borax, according to the intensity of the colour of the wine. (a.) The liquid remains wine lilac, or with some violet tinge, both elders, privet, whortleberries. Pass to P. (b.) The flu.id becomes bluish-grey-flax-blossom, gi'eenish or bluish-grey, with very faint trace of lilac, pure wine, whortle- berries, hollyhock, fuchsine. Pass to E. P. Treat a new portion of wine with sodium bicarbonate (as directed at I.). (rt.) The tint, at first lilac, changes afterwards to grey, slightly brownish, or to maroon. If anew portion be treated with sodium carbonate, according to C, and then heated to boiling, it becomes clearer, and loses its green tint. The lake obtained according to E. is deep blue-green. Dwarf Elder. (6.) The specimen remains grey, tinged with green, bottle- green, or yellowish. Sometimes (black elder) it acquires a lilac tint, which almost immediately disappears, changing to a greenish-grey-blue, whortleberries, black elder, privet. Pass to Q. Q. Treat a specimen of the wine with alum and carbonate of soda (as directed at E.). Shake the mixtui"e, and after a few moments throw it on a filter. (a.) The lake remaining on the filter is deep green-blue ; the filtrate is clear bottle-green. A sample treated with sodium carbonate (as at C.) darkens and becomes grey, slightly greenish, on heating to boiling. Black Elder. (6.) The lake is clear bluish or greenish. The filtrate is clear bottle-green. A sample treated with sodium carbonate (as at C), and heated to boiling, changes to dirty yellowish. Privet. (c.) The lake is ash-green faintly rose-tinted. The filtrate is bottle-green, with tint of maroon. A sample treated with sodium carbonate (according to C.) becomes deep grey on being heated to boiling. "Whortleberries. R. Treat a specimen of the wine with ammonia and ether, as directed at J. (rt.) The ether being decanted and evaporated, the fluid residue becomes rose-coloured on treatment with acetic acid. Fuchsine. {b.) The liquid left after the evaporation of the ether does not become red on acidification with acetic acid, natural wines, hollyhock, whortleberries. Pass to S. § 284.] COLOURING-MATTERS MIXED WITH WINE. 575 S. A sample is treated with its own bulk of a solution of aluminium acetate of 2° B. (a.) The colour of mixture remains winey, natural wines, whortleberries ; differentiate between them, as directed at L (a), and L (c). {b.) The colour of the mixture becomes violet-blue, holhjhoch, whortleberries. Pass to T. T. A specimen is treated with alum, and sodium carbonate (as at E.), and after a few moments filtered. (a.) The lake is clear green, slightly bluish, and rose-tinted ; filtrate is bottle-green, with little maroon. With borax (as at O), particularly if the sample has been concentrated, the liquid is grey with trace of lilac. 2 cc. of the liquid treated with 3 cc. of dilute ammonia (1 vol. of liq. ammonia with 10 vols, of water), and the mixture diluted with its own bulk of water, gives a liquid which is yellowish-grey, greenish or greenish-grey. The other characteristics as at L. Whortleberkies. (6.) The lake is green, slightly bluish, quite free fx'om rose, filtrate clear bottle-green. With borax the liquid is greenish blue-grey. With ammonia (as above), dark bottle-green. With aluminium acetate (as at S.), bluish-violet coloration. Hollyhock. Although somewhat difficult, this systematic method serves for the discovery of several colouring-matters mixed in one wine, if the indications of Tables A. and B. are carefully observed and followed. It is always desirable to determine the presence of fuchsine by the special reactions given further on. By means of Table B. the presence of one or several of the colouring- matters may be detected ; but before deciding, it is as well to verify by repeating, for the substances so found, the reactions of Table A. on the sample ; and also the more special characteristics afiven further on, for the identification of those substances. SPECIAL reactions FOR THE DETECTION OP CERTAIN OF THE COLOURING-MATTERS MIXED WITH WINES. Brazil Wood. — Even a very strong clarification (two or three times more albumen than mentioned at the head of Table B.) does not wholly decolorise the adulterated wine. It becomes yellow-buff, which on exposure to the air gradually changes to red. If a wine that has been adulterated with Brazil-wood is clarified, and then a skein of scoured silk, washed with dilute tartaric acid, be soaked in it for twenty-four hours, and then withdrawn, washed, and dried at 60° to 70^ the silk will be found to be dyed lilac-maroon, or red. In pure wine, the skein remains wine-colovired or lilac. 57G FOODS : their composition and analysis. [§ 284. If the dyed silk be now dipped into dilute ammonia, and heated to 100° for a moment, it becomes lilac-red, if Brazil-wood were present ; but deep grey, with scarcely a tinge of its original colour, if the wine were pure. If the ammonia be replaced by lime-water, the skein changes to ash-grey if Brazil-wood were present ; but to a dark, dirty-yellowish-red, if the wine were pure. Finally, if the skein be dipped into aluminium acetate, and then heated to 100°, it retains its wine-red lilac colour. This reaction differentiates Brazil-wood from logwood. Logwood. — If the colour due to logwood is in excess in the wine, ammonia gives it a shade of violet; if the proportion of logwood is small, the reactions B, L, N, of Table A., which are very delicate, should be tried. A skein of silk, prepared in the manner described for BrazH- wood, and treated with logwood, becomes dyed lilac-red, or maroon, which dilute ammonia changes to violet-blue tinged with grey, and which by acetate of aluminium becomes bluish-violet. Cochineal. — The lilac, or roseate tints due to the I'eactions A, B, H, K, of Table A., are very sensitive, the last being very characteristic ; the only substance likely to be confounded with it is the Phytolacca (Portugal berries), which is differentiated by the reaction B, of the same table. A skein of scoured silk, moixlanted with aluminium acetate soaked in the clarified wine for twenty hours is dyed of a wine violet colour, analogous to that of ])iire wine, on being dried at 100°. TJie colour does not change, even at 100°, by cupric acetate (exclusion of fuchsine) ; but if the skein be dipped into a dilute solution of zinc chloride, heated to 100°, and then wetted with sodium carbonate, washed with water and dried, the colour becomes fine purple, whereas with pure wine the tint would remain sombi-e grey-lilac. Cochineal may be discovered by the spectroscope if present in large quantity, but if it amounts to only about 12 per cent, of the total coloration, it cannot be so detected. It rapidly separates from wines, being precipitated in the lees. Fuchsine.- — This should be sought for in all wines found to be adulterated with other substances. The reaction J of Table B., p. 572, is very sensitive. Great care must be taken to avoid loss of rosaniline from imperfect decomposition of its salts in solution; moreover, arsenic should always be sought for where the wine is found to contain any aniline. Fuchsine rapidly separates from the wines to which it has been added. A skein of silk becomes dyed rose by soaking in a wine adulterated with fuchsine, and its colour passes to yellow on treatment with hydrochloric acid, but to bright red if the wine was pure. The dyed skein § 284.] COLOURING-MATTERS MIXED WITH WINE. 577 treated with dilute cupric acetate, and dried at 100°, becomes fine deep rose-violet if fuchsine is present, and of a lilac tinged with ash-grey if the wine is pure. This reaction is very- sensitive. Phytolacca. — (Portugal berries). The rose or lilac colorations of the reactions A, G, and especially C of Table A., are very sensitive. Hollyhock — (Althea rosea), much used. This substance imparts a peculiar flavour, which in a few months becomes actually dis- agreeable, while the colouring-matter itself rapidly precipitates. Beetroot. — This is generally employed only to mask other adulterants. The lilac tint of reaction C of Table A., if the beetroot is fresh, and the yellowish colours due to alkalies (reactions D, E and F of Table A.) are very sensitive, even with old decoctions. Black Elder, Dwarf Elder. — The dwarf elder imparts a faint turpentine odovir to the wines. The berries of both varieties are particularly used to communicate a special colour and flavour to port wine. The teinte de Fismes, (p. 565), largely used at Fismes, Paris, and elsewhere, is made by digesting 250 to 500 parts of eldei'-berries, and 30 to 60 parts of alum, with 800 to 600 parts of watei', and then submitting the mixture to pressure. M. Maumene reports having discovered as much as 4 to 7 grms. of alum per litre in wines adultei-ated with this substance. Sometimes (though rarely) the alum is replaced by tartaric acid. Wines adulterated with either yield a violet-blue lake (reaction H, Table A). By comparison with pure wine the difi"erence is very marked. A piece of flannel, or skein of silk, mordanted with aluminium acetate, heated for some time in the suspected wine, then washed, and immersed in water made faintly alkaline with ammonia, becomes green if the wine is pure, but dark brown if black elder is present. Probably the same reaction occurs with dwarf elder. Privet. — This is seldom used. The general reactions, particu- larly N and P of Table A., may be referred to (see also foot- note, p. 571). Kino. — Said to be coming into use. It is precipitated by gelatin, and givef, no definite absorption bands. According to Etti* kino-red is the anhydride of kinone, and has the formula C.sH,,On- Indigo. — The reactions A (5) and B (h) of Table B., p. 570, are so sensitive that they are alone sufficient to characterise indigo. Wool or silk mordanted with aluminium acetate, heated with 20 to 40 cc. of the suspected wine nearly to dryness, washed, and * Deut. chem. Ges. Ber., xi. , 1883, s. 1879. 38 578 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 285. then dipped into very dilute ammonia, become dirty green if the wine be pui'e, but blue if indigo be present. Indigo being often used to mask the too bright colours of cochineal and fuchsine, these should -always be sought for after the removal of the indigo by clarification with albumen. Indigo very rapidly separates from wines, and it may frequently be found in the lees, even when the wine itself gives no indication of its presence. Except in such cases as indigo and cochineal, it is only upon a series of concordant reactions that the presence of an artificial colouriiag-matter should be afiirmed. § 285. Mineral Substances, or Ash. — The ash of a great many wines, and especially of sheri-ies, imported into this country, consists nearly entirely of sulphates.* This is due either to sulphuring or plastering. It is found absolutely necessary to charge many wines slightly with sulphurous acid, some of which becomes sulphuric acid ; and in such a case the chlorides and carbonic acids are diminished in the ash, and the sulphuric increased, but the total weight of the ash itself is not materially increased. On the other hand, plastering (by which is meant the addition of plaster of Paris to the grapes before they are crushed) has the effect, by its reaction on cream of tartar, of producing a soluble sulphate of potassium, which may very materially increase the ash of the wine.f * The sulphuric acid in sherries ranges from 1 '5 to S grms. per litre (et^ual to from 19'0 to 93'S grains per bottle of -J gallon). t A plastered wine contains more potash than one not plastered, for in the latter tliere is a deposition and sei>aration of the hydro-potassic tartrate, but in plastered wines from double decomposition calcium tartrate is formed and deposited, whilst potassic sulphate passes into solution. According to Hilger, a plastered wine always contains more than '06 per cent., SO3, and shows a notable increase of the ash constituents. The general view of the reaction which occurs on the addition of calcic sulphate is that some tartaric acid is also set free according to the following equation : — 2C4H5KOG + SOiCa = C4H4CaOG + SO4K2 + CiHeOc, This free acid again acts on the potassic sulphate, forming SO4HK and C4H5KOC. R. Kayser, on the other hand, considers that free phosphoric acid IS formed-S04Ca + C4H5K(.)(., = SO4HK + C4H4Ca06, and then S04HK + P04H2K=:S04K2 + P04H3, and that it is this free phosphoric acid which gives the lively tint to red wines. Plastering clears a wine rapidly, because the calcic tartrate quickly separates. The main chemical changes, therefore, which can be traced are> briefly, a decrease in the tartaric acid and au increase in potash and sul- phates. The standard in use in the Paris Municipal Laboratory is '2 grms. per litre of potassic sulphate ; should a wine contain more than this quantity, it is considered plastered. The maximum amount of potassic Mulphate found by M. Marty in genuine wine was '6 grm. per litre. A ])reliminary examination of the wine is effected as follows:— A solution of 5'6US grms. of baric chloride and 100 cc. of HCl is made up to a litre with water. Two § 286.] COLOURING-MATTERS MIXED WITH WINK 579 Under absolutely normal conditions, the ash consists of carbon- ate, sulphate, phosphate, chloride of potassium, chloride of sodium, phosphate and carbonate of calcium, with very small quantities of magnesia, iron, silica, and frequently lithium and manganese. The ash from a litre of wine examined by Boussingault con- tained— Grms. Potash,* 842 Lime, 0-092 Magnesia, 0*172 Phosphoric Acid, 0-4 12 Sulphuric Acid, 0-09(> Chlorine, a trace Carbonic Acid, 0-250 Sand and Silica, Q-QOG 1-S70 With regard to the analysis of the ash, ikc, see p. 117, et seq. § 2S6. Detection of Fcuohorates and Fhtotsilicates. — Alkaline fluoborates and sometimes alkaline flnosilicates are used as antiseptics in the manu- focture of wine. Their detection is as follows : — 100 cc. of wine are treated ■with an excess of calcium hydrate and evaporated to dryness and ignited. Should fluoborate have been used the borate of lime is soluble ia acetic acid, while both fluoride and silicate of lime are insoluble in acetic acid. Therefore, after ignition the ash is treated with acetic acid and liltered. On evaporating the acetic acid solution to dryness the residue is tested for boric acid as described on p. 312. Silicates are determined in the insoluble portion of the ash remaining on the filter, and fluorides are detected by heating the ash so as to render it anhydrous, and, after mixing with a little sand, transferring it to a test-tube, then adding sufficient strong sulphuric acid to form a paste and closing the mouth of the test- tube with a cork carrying a small U-tube, a single drop of water having lieen put in the Ijend. On now heating the test-tube, fluoride of silicon is evolved ; but immediately decomposes on passing through the drop of w-ater ; hence there is a characteristic deposit of gelatinous silica in the U-tube. If, thei-efore, by these processes silicon and fluorine are detected, the amount of silica being in excess of what is usual in the ash, a silicofluoride has been added ; or, if fluorine and boracic acid have been found, this denotes the presence of a fluoborate. tubes are charged, each with 20 cc. of wine, and to the one is added !j cc. and to the other 10 cc. of the barium of chloride solution. If, after the pre- cipitate has subsided, the clear lir^uid from the 5 cc. tube gives no ])recipi- tate, the wine is not plastered ; or, if it gives a precipitate, whilst the second tube gives no precipitate, the plastering is beneath the standard ; but if the second tube gives a precipitate, the wine is plastered, and the usual methods of estimation must be adopted. * The rule is that nearly half the ash of a natural wine consists of K2O in combination. 580 FOODS : THEIR COMPOSITION AND ANALYSIS. BIBLIOGRAPHY. Batilliat. — " Traits surs les vins de la France. " Bekthelot et De Fleurien.— Jowrri. Chim. Med., 1864, t. x., p. 92. Cameron, C. A.— Analyst, i., 1887, p. 99. Chancel, G. — Comptes Rendus, Ixxxiv. 348-351. Chevallier. — Ann. d'Hyg., 1856, t. v.; 1857, t. vij. ; 1870, t. xxxiii. ; 1876, t. ij., &c. Ducleaux, ^.—Ann. Chim. Phys. [5], ij., pp. 233-289. DuPRE, A. — Proceedings of Soc. of Analysts, i., 1876; Analyst, i., 1877, 26, 186. Ferrikre, Sangle. — Art. "Vin" in Ch. Girard and A. Dupr6's "Analyse des Matieres alimentaires." Paris, 1894. Gautier. — Bull. Soc. Chim. [2], xxv. Griffin, J. J. — "The Chemical Testing of Wines and Spirits." London, 1872. Hager. —PAarai. Central Halle, 1872, No. 27. Hassall.— "Food and Air," 1874. Jacquemin. — Bull. Soc. Chim. [2], xxvj., 68-71 ; Analyst, i., 1877, 148. Las.saigne. — Journ. Pharm. et Chim., 4e ser., 1870, t. xi. ; Ann. d'Hyg. [2], 1831. Maumene, E. J.- — "Traite theorique et pratique du travail des vins." 2e ed., Paris, 1874. MoNAYON, Marius. — "La coloration artificielle des vins." Paris, 1890. Odeph, Alp. — Journ. Chim. M6d., 4e ser., 1860, t. vj. Neubauek, C. — "Ueber die Chemie des Weines." Wiesbaden, 1879. PA.STEUR. — "Nouvelles observations sur la conservation des vins." 8vo, Paris, 1868. PoGGiALE.— Jowrn. Pharm. et Chim. [3], 1859, t. xxxvj., p. 164. Schmidt, Conrad. — " Die AVeine des Herzoglich Nassauischer Cabineta- keiler." Berlin, 1893. Thudichum, J. L. W., and Dupre, A. — "A Treatise on the Origin, Nature, and Varieties of Wine." 8vo, London, 1872. VoGEL, H. VI.-Deut. Chem. Ges. Ber., ix., 1900-1911. PART VII. VINEGAE. § 287. Constituents of Commercial Vinegar. — Vinegar is a liquid resulting from the acetous fermentation of a vegetable infusion or decoction; it contains acetic acid, acetic ether, alcohol, sugar, gum, extractive matter, alkaline acetates, and tartrates, a variable amount of salts (depending on the substances from which it has been produced), and legally not more than 1*85 parts by weight of pure sulphuric acid per 1,000 of vinegar. Varieties of Vinegar. — The chief varieties of vinegar are as follows : — (1.) 2falt- Vinegar. — The great majority of commercial vinegars in this country are derived from the acetous fermentation of a wort, made from mixtures of malt and barley. ]\Ialt-vinegar is of a decided brown colour, in specific gravity varying from 1'017 to 1019 ; it is of various degrees of strength, the manufacturers distinguishing different kinds as Kos. 18, 20, 22, and 24, respec- tively, the last being the strongest, and containing about 4-G per cent, of acetic acid. That made by one of the largest firms in this country will be found to contain from "1 to -IG per cent, of combined sulphuric acid, and from -04 to -08 per cent, of chlorine, as chlorides. (2.) Wine-Vinegar is the chief vinegar in Continental com- merce. It is prepared from grape-juice and inferior new wines; that made from white wine is most esteemed. The wine-vinegars vary in colour from pale yellow to red ; they have nearly always an alcoholic odour; specific gravity from 1'014 to 1022. A litre of Orleans vinegar (according to Chevallier's* analyses of actual samples) saturates from 6 to 7 grms. of dry carbonate of soda. The extract from pure wine-vinegar varies from 1-38 to 3-2 per cent., the average being 1-93 per cent., and usually contains -25 grm. of tartrate of potash (see Table XLVa.). Vinegars of limited use are — Glucose-vinegar, recognised chiefly by the presence of dextrin, which may be precipitated by alco- hol ; beer-vinegar, from sour ale ; cider-vinegar, made both from apples and pears ; crab-vinegar, made from the crab apple, and used nearly all over Wales and Monmouthshire ; spirit-vinegar made from alcohol ; and various artificial vinegars. * Journ. iVHijrj., 1877, Ko. 45. 582 FOODS : their composition and analysis. [§§ 288, 289. TABLE XLVa. — The Composition op various kinds of Vinegar. .--§-15 i^"." 'S > -^ ill ill II < ^H '% Wiue Vinesar — p.c. p.c. p.c. Max., .... 10213 3-19 7-58 14-26 2-3 0-68 Mill., .... 1-0129 1-38 4-44 8-04 3-2 0-16 Mean, .... 1-0175 1-93 6-33 11-42 3-2 0-32 Spirit Vinegar- Max., .... 1-013 0-57 7-98 12-54 138 0-08 Min., .... 1-008 016 4-98 7-63 300 Trace. Mean 1-0082 0-35 6-34 9-86 18-1 0-04 Date Vinegar- Max., .... 1-0195 2-68 6-60 12-58 2-4 0-47 Miu., .... 1-0170 2-29 6 30 11-74 2-7 0-40 Mean 1-0185 2-40 6-44 12-10 2-6 0-44 Malt Vinegar, . 1-019 2-93 5-78 11-60 2-9 0-44 Malt Vinegar, which has also \ been made from sugar, . / 1-015 2-01 5-52 10-24 2 8 0-34 Rice Vinegar, 1-017 2-53 5-64 1099 2-2 0-34 Sugar Vinegar, . 1-010 l-6t 4-21 10-84 2-6 0-27 Vinegar certiiied to contain \ 70 p.c. of pyroligneous acid j 1-007 0-21 4-70 7-26 22-2 0-04 The above table contains results reduced and collated from Saiigle Ferrifere, Allen, Heliner, and others. § 288. Adulterations. — The adulterations of vinegar are — (1.) Water. (2.) Mineral acids, especially sulphuric, more rarely hydro- chloric, and still more rarely nitric acids. (3.) Metallic adulterations, or rather impurities ; such as arsenic* (derived from sulphuric acid), co])per,t lead, zinc, and tiu, from the solvent action of the acid on any metallic surfaces •with which it may have come in contact, (4.) Pyroligneous acid. (5.) Various organic substances, such as colouring agents, and capsicum. § 289. Analysis of Vinegar. — (1.) Water. — Vinegar should con- tain at least 3 per cent, of acetic acid (CoH^O,) ; a lower per- * " The observations of M. Deschamps induced us to anal}'se a vinegar sold by a certain Sieur C. . . . The i^resence of arsenic in this vinegar was ascertained, and the Sieur C. was compelled to confess that the vine^:^ar had been mixed with wood-vinegar. On resorting to the person who furnislied the latter product, the whole of the wood-vinegar in his possession was found arsenical, and seized, in order to be employed only for industrial use." — "Le Vinaigre," Chevallier, Journ. iVIIyrj., No. 46, June, 1877. t Seven out of twelve samples of vinegar sold in Paris, and analysed by Alfred Iviche, contained copper varying from 5 to 15 mgrms. per litre. Journ, Pharm. Chim. [4], xxvj. 23-28. 289.] 083 centage indicates dilution with water, for it is then so dilute as certainly not to be of the nature and quality of the substance usually sold as vinegar. The strength of vinegar may be accu- rately estimated by distilling 110 cc. until 100 cc. have been drawn over, that is, ten-elevenths. The 100 cc. will contain 80 per cent, of the whole acetic acid present in the 110 cc, and may be titrated ; or the specific gravity of the distillate may be taken, and the strength found from the followino; table : — er cent. Sp. gr. Per cent. Sp. gr. Per cent Sp. gr. 1 . . . 1-UOl 8 . . . 1-012 15 . . 1-0-22 2 . 1 -002 9 . . . 1-013 10 . . 1-023 3 .' '. . 1-004 10 . . . 1015 17 . . 1-0-24 4 . . . 1005 11 . . . 1-016 18 . . 1 025 5 . . . 1 -007 12 . . . 1-017 19 . . 1-026 (5 . . . 1 OOS 13 . . . 1-018 20 . . 1-027 7 . . . 1-010 U . . . 1-020 Vinegar may also be distilled in a vacuum produced by a mercury or water pump ; it should be distilled into caustic soda or potash of known strength, and then titrated back. By dis- tilling thrice to dryness, adding a little water each time, the whole of the acetic acid comes over. It will be necessary to test the distillate for the presence of hydrochloric acid, and also to take the acidity of the vinegar without distillation, so as to control the results. The titration of vinegar may be made with ordinary soda solution, and approximate results obtained.* If absolutely accurate determinations are required, it is best to add an excess of carefully weighed pure carbonate of lime to a known weight •of the vinegar; the liquid is boiled, filtered, and the residual carbonate of lime filtered off", dissolved in slight excess of normal hj'drochloric acid, and titrated back with caustic soda and cochi- neal solution. From the amount of carbonate thus found to have been unacted on by the vinegar, the total acidity is calculated. Carbonate of barium may with advantage replace the lime car- bonate. (2.) Mineral Acids. — A great many commercial vinegars con- tain no trace of free mineral acid ; and it has been amply shown that although about two-thousandth part of free sulphuric acid is legal, such addition is not by any means necessary for the preservation of the vinegar. The mineral acid, if present, is nearly always sulphuric, occasionally hydrochloric, and still more rarely nitric acid. Hydrochloric Acid is detected by the distillation already de- scribed, and the testing of the distillate with nitrate of silver. Nitric Acid may (in the absence of other reducing agents) be "■ The results are only approximate, because sodic acetate has itself a feeble alkaline reaction. 584 FOODS : their composition and analysis. [§ 289. detected by the rapid decoloration of a solution of indigo carmine added to the boiling vinegar, or the diphenylamine test (see article on Water). Sulphuric Acid cannot be detected by the usual chloride of barium test, for it fails to distinguish between free and combined sulphuric acid. The charring effect of the acid on paper, on sugar, or its action on starch (formerly taken as the basis of the older tests), is now replaced by more scientific methods, and need not be described here. One of the most speedy tests for the presence of mineral acids is that proposed by A. Hilger* : — Two or three drops of a solu- tion of methyl aniline violet (0"1 : 100) are added to 25 cc. of vinegar; if pure, no colour is produced; but if '2 per cent, of any mineral acid is present, the colour is blue; or if -5 per cent., blue- green; and if 1 per cent., green. Another useful test is that of M. Strohl;t it is based on the well-known fact, that oxalate of lime is insoluble in acetic, but soluble in mineral acids. The solutions requisite are — a solu- tion of calcic chloride (15'1 grms. to the litre) and a solution of crystallised ammonic oxalate (28-4 grms. to the liti'e); | cc. of each of these liquids is added to 50 cc. of the vinegar under examination, and if the turbidity which is at first produced does not disappeai-, the liquid contains less than — 1 '70 grm. per cent, sulphuric acid (specific gravity 1 "843) per litre. 2"S5 ,, hydrochloric acid ( ,, ,, 1'174) ,, 4-40 ,, nitric acid ( „ „ 1-174) The test, without claim to great accuracy, is extremely useful; for if any suspicious indication be observed, the vinegar may be then sxibmitted to a more elaborate examination for free acids. As speedy as any of the foregoing, and at once more scientific and accurate, is the process introduced by Mr. Hehner. Its principle is based upon the fact that vinegar always contains potash and soda salts of the organic acid ; hence, it is obvious that sulphuric or hydrochloric acids, if added in small quantity, merely decompose an equivalent quantity of acetate or tartrate, as the case may be, and as free acids immediately disappear; but if added in excess of the amount of acetates and tartrates, the excess remains as free acid. It thus follows, that if any undecomposed acetate or tartrate exists in the vinegar, it is. impossible for a free mineral acid to be present; and since the acetates and tartrates are decomposed by ignition into carbonates, * Archiv tier Pharmacie, 1876, 193. + Arch. Pharm. [5], 4, 342-346. § 289.] VINEGAR. 585 the readiest way to ascertain their existence is to examine the ash of the vinegar for carbonates. If that ash is neutral, free mineral acid is probably present ; if alkaline, no fi-ee acid can be present, although, of course, a small quantity may originally have been added. The qualitative test devised by Mr. Hehner is also made quantitative. If an accui-ately-estimated volume of d. n. soda solution is added to a known quantity of the vinegar, so as to neuti-alise slightly in excess the total amount of free minei-al acid present, on ignition the alkalinity of the ash gives the measure of the quantity of free sulphuric or hydrochloric acid. The exact details of this operation, as practised by Mr. Hehner, are as follows: — 50 cc. of the vinegar are mixed with 25 cc. of d. n. soda; the liquid is evaporated on a water-bath in a j^latinum basin, the residue dried at about 110°., and carefully incinerated at the lowest possible temperature — the ash need not be burned white. 25 cc. of a d. n. sulphuric acid solution are now added to the ash, the liquid heated to expel free COg, and filtered. The filter is washed with hot watei', litmus added,* and the acidity ascertained by d. n. soda. The volume of soda necessary for neutralisation directly gives the proportion of free mineral acid present in the vinegar, 100 cc. of d. n. corresponding to "49 grm. of HgSO^. If the amount of alkali originally added should have been insufficient, it is necessary to recommence the experiment. For this reason Messi's. Allen and Bodmer made some experi- ments in whicli the preceding manipulation was modified by neutralising the vjJiole of the acid, organic and inorganic, by soda solution. The results were satisfactory, but gi-eat care must be taken to titrate accurately. Another veiy satisfactory way of separating and identifying the free mineral acids in vinegar is the following : — Saturate a known quantity with cinchonine, evaporate to dryness, take up the quinine salts with spirit, recover the spirit by distilla- tion, dissolve the cinchonine salt in water, and precipitate by ammonia. The aqueous liquid will now contain the acetate of ammonia, together with the sulphate, chloride or nitrate ; if any one, or all three, of the free acids were present, the acids may be determined in the usual way. A method of separating free sulphuric acid from sulphates is to evaporate the vinegar to a syrup, precipitate the sulphates by alcohol, filter, wash the precipitated salts with alcohol, and determine the free sulphuric acid in the a-lcoholic solution. Provided suflicient alcohol be added, the separation of free from combined sulphuric acid is exact. * Instead of litmus, cochineal may be used ; the latter is unaffected by CO2, and therefore preferable. 5S6 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 289. Another method, the principle of which was proposed by Mr. Thresh, and which has been improved upon by Mr. W. C. Young, is to add to a known measure of vinegar an excess of BaClo; the chlorine in a portion of the liquid is now determined with great €are, the rest is evaporated, ignited, and the chlorine of the ash determined. The ditFerence represents the free mineral acid in terms of chlorine. The presence of free tartaric or citric acids quite invalidates the accuracy of the process, but, with these exceptions, it is generally applicable. (3.) Metallic Adulterations. — Metals in vinegar are detected by saturating the liquid with hydric sulphide, or by specially testing for arsenic, copper, zinc, tin, and lead, by the methods detailed in the author's work on " Poisons."* Metals of the first group may, however, be presumed absent, if there is no deepening of colour on saturation with hydric sulphide ; arsenic, if Reinsch's test gives negative results ; and zinc, if the nearly neutralised vinegar gives no precipitate with hydric sulphide. Interpretation of the Besults of Vinegar Analyses. — The chemists of the Municipal Laboratory, Pai'is, interpret the results of their analyses of wine-vinegar according to the following principles. The normal weight of extract and alcohol in wine is in the pro- portion of 4 : 1; a tenth per cent, of the extract is lost during acetification ; on the other hand, 130 grms., theoretically, of acetic acid is produced from 100 grms. of alcohol; in practice this must be diminished 15 per cent. The following table gives a, calculation on the above basis : — Alcoholic Strength. Acetic Acid. Extract of Vinegar. Per cent. (vol.). Per cent. Per cent. G 5-315 ros 7 6211 1-26 8 7-105 1-44 9 8-00 1-62 10 8-895 1-80 11 9-801 1-98 12 10-71 2-16 The relation existing between the acetic acid and the extract is equal to 4-9 ; if a vinegar has a relation higher than this, allowing at the most a tenth (that is, really 5-0), an addition of alcohol is indicated, whether an actual addition or Avhetlier from the use of a fortified wine. The table also indicates the maxi- mum and minimum quantities of acid which can be furnished by * Poisons : their Effects and Detection, 3rd ed. Loud. 1895. § 2S9.] VINEGAR. 587 wines of from 6 to 12 per cent, of alcohol ; and should the pro- portion of acetic acid fall below 5-3 per cent., watering may be concluded, especially if the potassic tartrate, ash, and extract are diminished in tlie same proportions. The French standard for vinegar is, therefore, near the standard of our own pharmacopeia (5-41 per cent.). It becomes of importance to distinguish the different vinegars from one another — that is, to be able to say, at least, whether the vinegar is derived from the fermentation of glucose, from wine, from malt, or whether it is derived from pyroligneous acid or is a diitilled product. The table on p. 582 will assist the analyst in coming to a conclusion. Wine- vinegars always contain some alcohol and this may be estimated by careful neutralisation of the vinegar and distillation ; wine-vinegars also contain bi- tartrate of potash (from 0-65 to 0-36 per cent.); often small quantities of reducing sugar (fi-om -007 to 0*46 per cent.) as well as some of the wine ethers are present ; the relation between extract and acid as a rule does not exceed 3 '5. Malt- vinegars also contain some alcohol, but are distinguished from wine- vinegars by the absence of bitartrate of potash. The phosphoric acid ranges from about 0'05 to 0*1 per cent. Grain -vinegars also contain a fair amount of nitrogen ; this, according to Alien, is a valuable critei-ion. The nitrogen, calculated as albuminoids, ranges from 06 to 0*7 per cent. Vinegar made from sugar gives low albuminoids. Pyroligneous acid before being made into vinegar is usually distdled, and, therefore, necessarily has a low extract, a low ash, and may, rarely, contain some tarry products. It may be recognised according to the French chemists by the larger amount of furfurol than is the case with other vinegars. Occasionally sulphate and acetate of potassium are fomid in pyroligneous acid. A vinegar made from glucose, or from rice converted into glucose by inverting with sulphuric acid, may show traces of its origin in increased sulphuric acid and sulphates by concentrating and adding strong alcohol ; dextrin may also be precipitated and be identified. It has been proposed by Hehner to calculate the various con- stituents on the original solids of the vinegar. As 60 parts of acetic acid are theoretically produced from 90 of glucose, the acetic acid is multiplied by 1-5, as representing theoretically the sugar from which the acetic acid was derived. To this figure is added the total extractive matter still contained in the wort- vinegar, and the number thus obtained represents the "original solids '"' of the wort. Examples of this calculation are shown ia Table LXVa. This number obtained, the ash, acetic acid, and 588 FOODS: their composition and analysis. [§289. albuminoids may be calculated out in terms of the original solids, which is doubtless a better guide to composition than the usual way of stating them. For example, calculating a malt-, a "spirit-," and a " sugar "-vinegar in per cent, of the total solids, the results are as follows : — Acetic acid, Albuuiinoids, . Phosphoric acid, Ash, . ilalt- Spirit- Sugar- r'inegur. Vinegar. Vinegar. 49-8 63-2 50 5-9 0-8 1-15 0-8 0-0 0-18 3-8 0-4 3 35 Vinegar is often coloured with caramel ; sometimes simply acetic acid thus coloured is sold as "vinegar." The estimation and detection of caramel has been worked out by C Amthor"*' as follows : — 10 cc. of the liquid are put into a flask with 30 to 50 cc. of paraldehyde, and absolute alcohol added until the liquids have mixed thoroughly. The flask is corked and allowed to remain \indisturbed for twenty-four hours ; by that time the caramel will have precipitated. To still further identify it, the liquid is decanted, the precipitate washed with a little alcohol, dissolved in hot water, and reduced by evaporation until it measures a cubic centimetre. On now treating the solution with phenylhydrazine, an insoluble compound is obtained. This is distinguished from the osazones of sugar by its amorphous appearance. Estimations of caramel are best made by imitating the colour of the paraldehyde precipitate ; for this purpose the precipitate is dissolved in water, made up to 100, and a solution of known caramel content used as a comparison liquid. Possibly the phenylhydrazine compound could be weighed, but the limits of accuracy in this direction have not yet been defined, and the formula of caramel is not accurately known. A. Sabaneef and J. Antushevitchf believe, from cryoscopic researches, that its formula is Ci^sHiggOso. * Zeit. f. anal. Chemie, xxiv., 30-33. . t/. Buss. Chem. Soc, xxv., 23-31. BiBLioGRAPnr. 589 BIBLIOGRAPHY. Allejj, a. H., and Bodmek, R. — Experiments on the Determination of Free Acids in Vinegar. Analyst, iii. , 1878, 268. On Vinegar. Analyst, Jany. and Feb., 1894. BoETTGER. — Journ. Pkarm. et Chim. [3], 1845, t. viij., p. 11,3. Brown, J. Adrian. — The Chemical Action of pure Cultivations of Bac- terium aceti. J. Cheiyi. Soc. {Trans.), 1886, xlix. 172. On an Acetic Ferment which forms cellulose. Jb., 432. Cauvet. — Examen et analyse des vinaigres. Ann. d'Hyg. et de Med Lig., 2e serie, 1874, t. xli. Chevalliek, Gobley, et Journell. — Essais sur le vinaigre, ses falsifica- cations, les moyens de les reconnaitre, \o. lb., 1843, t. xxix., p. 55. Chevallier. — lb., 1839, t. xxi., p. 86. Journ. d'Hyg., 1887, No. 45. Ferriere, Sangle. — "Vinaigre" in Girard and Dupre's "Analyse des Matieres Alimentaires." Paris, 1894. Oaultier de Claubky. — Ann. d'Hyg. et de Mid. Leg., 1841, t. xxvij. Guibourt. — Journ. Fharin. et Chim. [3], 1846, t. x. , p. 91. Hehner, 0. — Sulphuric and Hydrochloric Acids in Vinegar. Analyst, i., 1877, p.JOo. Pasteur. — "Etudes sur le vinaigre." Paris, Svo, 1868. Thresh, J. C. — Pharm. Journ., July, 3, 1875. Young, W. C — On Sulphuric Acid in Vinegar. Analyst, ii., 1878, 163» 203. 590 FOODS : THEIR COMPOSITION AND ANALYSIS. [^§ 290, 291. LEMON JUICE AND LIME JUICE. §290. Lemon juice is the juice of the Citrus linionum. and lime juice that of the Citrus acida and Citrus linietta. The Board of Trade standard for lemon juice is a density of 1 '030 [when dealcoholised], and an acidity equivalent to 30 grains per ounce of citric acid. The British pliarniacopceia directs that lemon juice should have a specific gravity of L039, and should contain 32'5 grains of citric acid per ounce. Lemon, lime, and bergaraot juice are all similar in their composition, containing citric acid as the predominant free acid, and small cpiantities of acetic, formic, and other organic acids, together with albumen, sugar, mucilage, and extractive matters. The mineral matter is very small, and contains 54 per cent, of its weight in potash and 15 percent, of phosphoric acid. The juice is expressed in England, and also in Sicily, -where the method of preparation is to mix one ounce of brandy with ten ounces of the juice, and over the surface of the liquid to pour a layer of olive oil. This crude process of preservation is eiiectual, but is now being supplanted by more modern methods. The following table shows the specific gravity, free acid, and combined organic acid of the citric commercial juices, the acid being expressed in terms of crystallised citric acid [^AUen^ : — Specific Gravity. Free Acid. Ozs. per gallon. Combined Or.ijanio Acid. Per gallon. Lemon Juice — Eaw Sicilian, ,, English, Concentrated, Berganiot Juice — Concentrated, Lime Juice- Raw, . Concentrated, 1-04 to 105 1-20 to 1-25 1-22 to 1-25 1-035 to 1-040 1-28 to 1-38 6 to 9 11 to U 57 to 72 47 to 55 10-6 to 13-5 82 to 112 0-85 0-3 6 to 8 7 to 8 0-4 to 0-7 8-6 § 291. Adulterations and Analysis of Citric Juices. — Lime juice has been rather extensively adulterated, and at the present time it is by no means uncommon to meet with a wholly fictitious article under this name. Sulphuric, hydrochloric, and nitric acids are the main foreign ingredients to be sought for. The general principles of the detection of these acids in a free state are entirely tlie same as in § 292.] LEMON JUICE AND LIME JUICE. 591 vinegar, and the remarks with regard to the allvalinity of vinegar-ash when genuine apply equally to the ash from the citric juices. (See p. 584.) Good juice contains insignificant traces of sulphates and chlorides, so that the mere addition of silver nitrate or of barium chloride will at once show whether there has been any tampering with the liquid. Nitric acid may be detected by the ordinary tests for that acid. The juice may be boiled with metallic copper, when red fumes will appear, should nitric acid be present. Or. it may be much diluted and filtered, and one portion be made neutral with burnt magnesia, and boiled to expel all free ammonia ; afterwai-ds, by acting iipon the liquid with a copper zinc couple, any nitrates may be turned into ammonia, distilled over, and titrated in the usual way. F. Scribani* adds to the suspected lemon juice an aqueous solution of ferrous chloride, strongly acidified with hydrochloric acid and free from ferric salt. The liquid is boiled for a few minutes, and then a little sulphocyanide is added. If nitric acid has been present, it will have oxidised the ferrous salt into a ferric salt, and a deep blood-red colour will be produced by the^ test. § 292. The Analysis of Lime Juice, with a view to ascertain its sti-ength, is confined to the determination of the amount of citric acid and citrates. The amount of citric acid is determined frequently for technical pui-poses by the aid of a special hydi'ometer, called " a citro- meter," but this method is not exact enough for the purposes of the food-analyst. Nevertheless, it will always be well to take the specific gravity of the juice by a sjiecific gravity bottle, or by any other reliable method, first boiling off" any alcohol which the juice may contain, and making it up to the same bulk as befox-e the dealcoholisation. The amount of free acid may be estimated by means of deci- normal soda. A known quantity of the juice is taken and coloured with phenol-phthalein, and then d. n. soda is run in until the colour changes. About 10 cc. of the ordinary raw juice may be taken and diluted to 50 cc. ; but the concentrated juice must be much diluted before titration. The free acid known, the next step is to determine the amount of citrates and other organic acids combined with bases. For this purpose, the measured quantity of the juice which has already been neutralised by soda, is evaporated down, charred, and the charred mass treated with a known volume of decinormal *Gaz. CJiim. Ital, viii. 294 592 FOODS : their composition and analysis. [§ 292. sulphuric acid, which must be sufficient to more than neutralise the carbonates. The acid solution is filtered and neutralised by d. n. soda; this will give the necessary data from which to calculate the amount of sulphuric acid used by the carbonates produced by the action of heat on the organic acids. This amount is equi- valent to tiie total amount of organic acid ; if expressed as citric acid, forty-nine parts of sulphuric are equal to seventy of SHgCgHgOyHgO, or sixty-seven of H„C,-Il^O^Iip. The amount of free acid already obtained is now subtracted from the total acid, the difference being that Avhich is combined with bases. To ascertain the amount of real citric acid present in the juice, it must be determined as citrate of lime, for it need hardly be said that the process given above does not distinguish between nialates, meconates, or any other organic acids converted by heat into carbonates. To determine the real citric acid, Mr. Warington* recommends the following process: from 15 to 20 cc. of raw lime juice are exactly neutralised by d. n. soda, the whole made up to about 50 cc, and heated to boiling. While boiling, so much of a solution of chloride of calcium is added as is known to be rather more than equivalent to the total amount of organic acids present. The boiling must be continued for half an hour, and the precipitate collected and washed with hot water. The filtrate and washings are evaporated to a small bulk (not more than 15 cc), and a little ammonia added to exact neutralisation, if the liquid gives an acid reaction ; a farther precipitation takes place, and this second precipitate must be collected on a filter. The filters are dried and burnt up at a low heat, and their neutralising power with regard to acid is determined. For this piirpose, the ash may be dissolved in standard hydrochloric acid, and titrated back ; each cc. of normal HCl neutralised is equivalent to -070 grni. of crystallised citric acid. If either oxalic or tartaric acids should be present, the results are, of course, inaccurate. *Journ. Chem. Soc, 1875, 934. PART VIIL-CONDIMENTS: MUSTARD, PEPPER, &.C. MUSTARD. § 293. Mustard is made from the seeds, finely ground, of the Sinapis nigra, or black mustard, or from those of the Slnapis alba, or white mustard, or again, from a mixture of both varieties. The manufacturer i-educes the seeds to powder, and passes the product through' a series of sieves. The portion in the first sieve is called the dressings, that which passes through is an impure mustard flour. The impure tiour, on being passed through a second sieve, yields the pure flour of mustard and a second quantity of dressings. The dressings are utilised, by being sub- mitted to pressure, for the sake of the fixed oil they contain. Microsco2ncal Structure of the Seed. — The white mustard seed is made up of the husk and the seed proper. The seed proper is simple in structure, consisting entirely of minute oil-beaving cells; their size averages -00041 inch in the finely powdered seed ; and they look extremely like starch corpuscles, but neither pol- arise light nor strike a blue colour with iodine. The complicated structure of the husk of the mustard seed, in part unravelled by Hassall,has more recently been fully investigated by the labours of Sem- plowskis and v. Hohnels.* It is built up of no less than six layers (see fig. 67). (1.) The most superficial is composed of almost quadratic thin-walled cells (•05 — 1 micro-millimetre in diameter) covered with a thin cuticle. The lumen Fig. 67.— Section through the coats of the seed of Sinajns alba. — sk, The outer skin, filled with swollen mucus ; p, the polygonal cells in section, their form better seen in fig. 6 into certain drinks), that it is absolutely necessary to be acquainted with their chemical composition. Both varieties of almond agree in containing about 50 per cent, of a bland flxed oil (consisting chiefly of olein, and liable to become rancid), as well as an albuminous principle, emulsin, sugar, gum, and woody fibre ; but ouly in the bitter almond is found, in addition to the fore- going, aniygdalinr § 308. The Oil of Almonds is a thin fluid oil, of a clear yellow colour, specific gravity 0-914 to -920, not coagulated by a cold of — 10° ; at — 16° it becomes cloudy, and at — 22° it solidifies to a white butter. Oil of almonds appears to be rather frequently adulterated with other oils. 2 5 drops of the oil, shaken with an equal bulk of nitric acid (specific gravity 1-20) and bisulphide of carbon, should not show any colour after standing a few minutes ; if it becomes within half an hour yellow, or reddish-yellow, the change indicates oil from cherry or apricot kernels. The following test will detect drying oils : — Dissolve one part of starch in 3 parts of warm nitric acid, of 1*20 specific gravity, and warm in a capacious vessel over the water-bath with 10 parts of almond oil, until all evolution of gas ceases. The oil after cooling is within two days changed into a warty, crystalline, greasy mass of elaidin. Should it, however, contain a drying oil (poppy, for example), it either remains quite fluid or semi-fluid, according to the proportion of the adulterant present. The colour of the elaidin is also a guide; that produced by the sweet almond is pure white, by the bitter, yellowish-white, and by the small or inferior kinds of almonds, brownish-yellow; if the elaidin 618 foods: their composition and analysis. [§309. shoiild be red, it denotes adulteration of some foreign oil, especially of sesame. Pure almond oil dissolves in 25 parts of cold and 6 of hot alcoliol. The above tests, and in addition the low temperature required for congelation, should detect all ordinary adulterations. § 309. Amygdalin (C^oHo^NO^^) is a glucoside, discovered in 1830 by Eobiquet and Eoutrou-Charlard. It may be extracted from almond-cake by boiling alcohol of 95 per cent., and then precipitated from the somewhat concentrated alcoholic solution by ether. Amygdalin* crystallises from 80 per cent, alcohol in colourless glittering scales, containing two atoms of water : it can also be obtained in crystals. Amorphous amygdalin of the before-mentioned cherry-lavirel leaves and buckthorn bark is best obtained by the following method : — The dried buckthorn bark is boiled with absolute alcohol, agitated with lead oxide, and evaporated to dryness. Dried in a vacuum over SO^H^ it forms a brittle, yellow, transparent, resin-like mass, which, when heated to 100°, becomes dark-brown; it can be dissolved by boiling alcohol and by water, but is insoluble in ether. Although amor- phous, it is a crystalloid with three atoms of water, as proved by dialysis from water or weak spirit, but in such a case it loses one atom if dried over sulphuric acid. At 100° to 120° it may be obtained anhydrous. Amygdalin possesses no smell ; it has a slightly bitter taste ; its reaction is nenti-al, and it polarises to the left [a] r = - 35-57°. It dissolves in all proportions in boiling water, and in 12 parts of cold of 10°; requires 148 parts of alcohol, specific gravity 0-939, 904 parts of alcohol, specific gravity 0-819, if cold — but if boiling, 11 parts of the first and 12 of the last ; it is insoluble in ethei'. It melts at 120°, and begins to carbonise at 160°, when it develops a caramel smell, and is at length fully destroyed. * Lelimann, in his recent elaborate researches, found the method of Liebig and Wohler the best for obtaining crystalline amygdalin. The process con- sists in boiling the substance with strong alcohol (of 04 to 95 per cent.) twice successively, after having first removed the fixed oil by petroleum benzine, concentrating to about one-half or one-sixth of its volume ; and then adding ether, which precipitates the amygdalin, and removes any of the remaining fixed oil. Lehmann obtained from Bitter almonds, . 2-5 per cent, crystallised amygdaUn.. Cherry-kernels, . . 0-82 „ Plum-kermels, . . 0-96 ,, ,, ,, Apple-seeds, . . . 0-60 ,, ,, ,, Peach- kernels, . . 2-35 ,, ,, ,, Cherry-laurel leaves, . 1 -38 per cent, amorphous amygdalin. Barli oi Jihamnusf rang ula, O'T ,, ,, ,, Both of these latter substances contain hydrocyanic acid ready formed. § 309.] THE SWEET AND BITTER ALMOXD. G19 Amygdalin, by the action of dilute liydrochloric acid, splits up into glucose and mandelic acid, volatile oil of almonds, and formic acid. If boiled with solutions of potash or baryta it forms ammonia and amygdalic acid. The most interesting decomposition is, how- ever, that which takes place by the action of emulsin ; it then breaks up into volatile oil of almonds, hydrocyanic acid, and formic acid. Volatile Oil, or Essence of Almonds, does not exist as such in the bitter almond ; it is, as above explained, the result of the decom- position of the amygdalin. The oil of almonds, when properly pui-ified from prussic acid, is identical with the hydride of benzole, CyH-OH. It is colourless, thin, turning a ray of polarised light to the right, of a peculiar, pleasant odour, and a burning aromatic taste. Its specific gi-avity is 1-04.3 to 1-07, usually TOG {Uirsch). Its boiling point is ISO"". By the action of light and air it is gradually oxidised into benzoic acid. It is soluble in eqiial parts of alcohol, 0"S30 specific gravity, and in about 30 parts of water. The ethereal or volatile oil is officinal in the French, Swiss, and Norwegian pharmacopoeias. The ethereal oil is much adulterated. The analyst will specially look for alcohol, prussic acid, nitro- benzine, and ethereal oils. If alcohol-free, the addition of an equal weight of fuming nitric acid produces no effervescence, and after two or three days the mass becomes emei'ald green, and crystals of benzoic acid appear. On the other hand, if it contain alcohol from. 0-08 per cent, upwards, there is immediately a strong effer- vescence. Some of the tests given for alcohol at pp. 470, et seq., will be of sei'vice. The detection and estimation oi prussic acid in the essence is carried out on the principles detailed in the article on Prussic Acid in the author's work on "Poisons." Nitrohenzine is indicated when the essence is not entirely soluble in a solution of bisulphate of potash, and the specific gravity is higher than 1-07, the specific gravity of nitrobenzine being 1-20 to 1-29; the boiling point will also be raised. In such a case nitrobenzine should be specially tested for, by changing it into aniline by reducing agents. For this purpose 10 parts of dilute sulphuric acid (specific gravity 1-117) may be added to 10 of granulated zinc and 1 part of the essence. At the end of two hours (after frequent agitation) the fluid is passed through a moistened filter, and a crystal of chlorate of ])otash added to the filtrate with a drop of concentrated sulphuric acid. If a violet or red colour is produced, it is due to the presence of an aniline salt, produced from nitrobenzine ; but if there is no colouration, nitrobenzine must have been absent. 620 foods: their composition and analysis. [§309. Another special method nsed for the detection of nitrobenzine ■was proposed by Maisch : — 1 grm. of the essence is dissolved in twelve times its volume of alcohol, -75 of caustic melted potash is added, and the whole heated until the liquid is diminished to about one-third. The pure essence, on cooling, is of a light brown colour, and dissolves entirely in water; but if nitroben- zine is present, the residue is brown, crystalline, and insoluble in water. The action of sodium on the essence may also be utilised as a test : — Pure almond essence, when treated with sodium, gives white flocks; if nitrobenzine should be present, the sodium is immediately covered with yellow or brown flakes, according to the amount of adulteration; if the percentage rises as high as 0-30 to 0-50, the whole liquid after a minute becomes thick and opaque. {Dragendorff.) However, the action of potash alone on a sample adulterated ■with nitrobenzine is tolerably conclusive. If one grm. of the essence is treated in a test-tube with half its weight of pure caustic potash, a yellow coloration is produced, should the essence be pure; but if nitrobenzine be present, the tint soon becomes yellowish-red, and at the end of a minute green. On the addition of a little water, the mixture separates into two layers, of which the lower is yellow and the upper green, the latter changing in the course of a day into red. Most foreign ethereal oils may be detected by the bisuljjhate of soda test: — If a little of the pure essence be dropped into a warm solution of this salt, of from l'24r to 1'26 specific gravity, shaken, and then diluted with hot water, it is fully dissolved; other essences, on the contrary, are insoluble. BIBLIOGRAPHY. Bette. — Ann. Chem. Pharni., xxi. 211. BocTRGOiN. — Falsitication de I'esseuce d'araandes amSrespar le nitrobenzole ? Joarn. Pliarm. et Chhn., 4e serie, 1872, t. xv., p. 281. Flttckiger.— iV^. Jahrb. Pharm., xxxiv. 202; Testing of Bitter Almond Oil, Pharm. Journ., Oct., 1870, p. ;^21. HinscH— "Die Prufnng der Arzneiinittel." Berlin, 1875. HusKMANN.— "Die Pilanzenstofte." Berlin, 1871. Lehmann. — Proc. Amer. Pharm. Assoc, 1875, 437. Maisch. — Falsification de I'essence d'amandes ameres. Journ. Pharm. et Chhn., 3e se'rie, 1858, t. xxxiv., p. 75. Eedwood.— Means of recognising the Puritj' of Essence of Almonds. Pharm. Journ., and Joimi. Pharm. et Chim., 1858. EoBiQUET and Boutron Ciiarlard.— J«w. Chem. Phys. [2], xliv. 352; Journ. Pliarm., [2], xxiii. ; Ann. Chem. Phaiin., xxxi. 211. §§310,311.] ANNATTO. 621 SouBEiRAX. — " Dictionnaire des Falsitications." Paris, 1874. Van den Corpit. — Sur la falsification de I'essence d'amandes ameres. Journ. Chim. Med., 4e serie, 1855, t. i., p. 585. WiCKE. — A7m. Chem. Pharm., Ixxix. 79; Ixxxi. 241; Ixxxiii. 175. WiNCKLER. — Rep. Pharm., Ixv. 1. WoHLER and Liebig.— -4ren. Chem. Pharm., xxii. 1. ANNATTO. § 310. Annatto is a colourinf]f-matter obtained from the seeds of the Bixa orellana, chiefly prepared in Brazil and Cayenne. Although not vised itself as a food, it enters into several articles of consumption, and has been employed to colour milk, butter, and cheese. Microscojncal Characters. — When annatto is examined by the microscope, the outer red portion presents an almost homogeneous appearance, and the surface of the seed proper consists of narrow or elongated cells or fibres vertically disposed, while the inner white portion consists of cells filled with starch corpuscles, well defined, of medium size, and resembling in the elongated and stellate hilum the starch granules of the pea and bean. In manufactured unadulterated annatto, but little structure is met with. Portions of the outer cells may be seen; and in those specimens, which in the course of their preparation have not been subjected to the action of boiling water, a few starch granules may be noticed. Since this is the case with annatto itself, we can the more easily detect the presence of most foreign vegetable substances, such as turmeric powder, the starch of wheat, rye, barley and sago flours. The salt and alkali present in the annatto generally greatly alter the appeai'ance of the turmeric. Most of the colouring-matter of the cells is discharged, so that the starch corpuscles contained within them become visible. Loose starch granules of turmeric may also be frequently seen, and in conse- quence of the action of the alkali much enlarged. § 311. Chemical ComiMsition of Annatto. — Dr. John found the I)ulp surrounding the fresh seed to consist of 28 parts of coloui-ing resinous matter, 26-5 of vegetable gluten, 20 of ligneous fibre, 20 of colouring extractive matter, 4 formed of matters analogous to vegetable extractive, and a trace of spicy and acid matters. The colouring-matter consists of a red substance — hixin, associated with a yellow, orellin; the latter has been as yet but little studied. 622 foods: their composition and analysis. [§ 312. Bixioi, CjjH^gO^, when pure, is an amorphous, resinous, cinna- bar-red substance. It is scarcely soluble in water and bisulphide of carbon, soluble in about 89 parts of cold, and in 25 o'i hot, alcohol; in 345 parts of ether, in 93 of chloroform, and also in the caustic and carbonated alkalies. The alcoholic solution is coloured orange-red by lead acetate, brownish-yellow by chloride of mercury and acetate of copper, brown-red by chloride of iron, and it is precipitated yellow by stannous or stannic chloride; concentrated sulphuric acid produces a deep blue. A solution of bixin in an alkaline liquid, on neutralisation with an acid, gives a precipitate of the resin, and in this way it may be purified. § 312. Adulterations. — Annatto is one of the most adulterated stibstances met with in commerce, the adulterants being both organic and inoi'ganic. The oi'ganic substances used are — tur- meric, rye, bai-ley, and wheat flours. The inorganic — sulphate of lime, carbonate of lime, salt, alkali, an oily substance (probably soap), red ferruginous earths (mostly Venetian red), red lead, and copper. When large quantities of flour and lime are used, the colour of the annatto is so reduced that it becomes necessaiy to use salt, alkalies, and the red earths, to restore it to its original standard. Salt heightens the intensity of vegetable reds, hence its use. Lead is probably introduced into the annatto thi-ough the Venetian red iised. Copper is added to prevent the annatto becoming attacked by fungi. The following is an analysis by the writer of a fair commercial sample. No. 1. The sample was in the form of a paste, colour deep red, odour peculiar, but not disagreeable : — Water, 24-2 Eesin 28-8 Ash 22-5 Starch aud Extractive Matter, . . . . 24-5 100 The following is an analysis of an adulterated sample. It was in the form of a hard brown cake, texture hard and leathery, odour disagreeable : — Water, 13-4 Resin, ll'O Ash— consisting of Iron, Silica, Chalk, Alumina, 1 .^.o and Salt, J- 48 d Extractive Matter 27-3 100 §§ 313, 314.] OLIVE OIL. 623 § 313. The Analysis of Annatto, as may be gathered from the preceding description, principally resolves itself into a deter- mination of the ash and an estimation of the resin. The former is determined in the usual way, the latter by exhausting the sample by boiling alcohol, getting rid of the spirit by evapora- tion, and then redissolving the extract thus obtained in an alkaline solution, and finally precipitating the nearly pure resin by careful neutralisation with an acid. BIBLIOGRAPHY. Bltth, a. Wynter.— " Diet, of Hygiene." London, 187G. BoLLEY and Mylius.— C/ie?n. CentralhL, 1865, 40. Cooley's " Dictionary of Practical Receipts." London, 1872 John. — Cliem. Schri/t., ii. 73. Stein, W. — Journ. pract. Chan., also Chem. CentralhL, 1867, 812. OLIVE OIL. § 314. Olive oil is derived from the fruits of the olive tree^ Olea Europcea sativa, of which there are many varieties ; in Italy alone, three hundred. The composition varies within certain limits according to the species and locality, and as to whether the oil has been expressed or extracted by solvents. The finest edible oils are known, under the names of "virgin oil," "Provence oil," "Aix oil;" and a second quality is sold under the name of " finest Tuscan, cream,"' Olive oil contains about 28 per cent, of solid glycerides consisting of palmitin, stearin, and a small quantity of arachin. The liquid portion of the oil contains olein, linoleic acid, and unsaponifiable matter (cholesterol) in the following proportions : — Olein, 66 per cent. ; linoleic acid, 5 per cent. ; unsaponifi- able matter, 1 per cent. Most specimens of olive oil when examined by the spectro- scope show a chlorophyll spectrum. The constants of olive oil, as compared with others, are set forth in the following table : — 624 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 314. o p o o S o Maumen(? Thermal Test. 34° to 35°, mean value •35. 75° to 80°. 44° to 67°. 49° to 51°. 63° to 64°. 79° to 86° cocoa-nut olein. 26° to 27° (Allen). 86° to 88°. II 79 to 88 100 to 117 87 to 100 97 to 102 103 to 112 llltollO 8 to 9 130 to 138 180 to 196 191 to 190 190 to 190 175 to 179 188 to 191 189 to 190 246 to 268 190 to 195 94-90 96-0 96-0 95-0 96 96-0 88-0 95-4 .B .5 i 1 2 1 21° to 22° 31° to 36° 24° to 30° 17° to 18° 18° to 22° 14° to 16° 16° to 20° 15° to 16° (S -2° 0° to - 1° -3° to -7° - 2° to - 10° - 4° to - 6° -10° to -15° 16° to 20° -18° I 1 < •S62 •867 to •80S •864 to •869 •856 to -864 •868 to -871 •868 to -869 •873 •873 < •914 •922 to -930 •916 to ^922 •914to^917 •921 to ^924 •921 to ^922 •9-23 •923 to ^924 Olive oil, . Cotton-seed oil, Aracliis oil, Rape oil, . Sesame' oil, . Maize oil. Cocoa-nut oil, . Poppy oil. .2 o •Sb a fe pi § 315.] OLIVE OIL. TABLE L.— Mixed Fattt Acids. 625 Melting Point. Saponification Value, Mgrms. KHO. Iodine Number. Olive oil, . Sesame oil, Eape oil, . Cotton oil, Arachis oil. Poppy oil, . 24° to 27° 24° to 26° 16° to 19° 30° to 36° 28° to 31° 20° to 21° 192 199 170 to 177 204 to 208 199-7 199 85 to 90 109 to 112 99-8 to 103 111 to 113 96-5 to 103 9G to 103 § 315. Adulterations of Olive Oil. — Olive oil commands a good price, and is therefore extensively adulterated -with other oils, especially with sesame, rape, cotton, arachis, and poppy oils. Adulteration is indicated by one or more of the " constants " deviating from the normal. Refraction of Olive Oil. — Jean's refractometer may be applied for the detection of certain adulterants. Oliveri gives the following values : — Deviation. Olive oil, 0° to 2° Sesame oil, . . . . . . . 15°'5 Arachis oil, ....... 7°"5 Cotton-seed oil, 18° Colza oil, 26°-5 Poppy-seed oil 28° -o Castor oil, 4r-44 J. H. Long* has examined a number of oils and determined the specific gravity at 24° as compared with water at 4°, the oils being weighed in a vacuum. He has also determined the refraction at 20°, using the sodium light ; the following are his chief results : — Olive oil. Cotton-seed oil, Sesame oil, . Mustard oil, Castor oil, . Lard oil, Peanut oil, . The variations caused by temperature he found to be 0-00068 specific gravity, and 0°'0004 refraction, for each rise of 1 degree in temperature. Refraotiv Specific Gravity. Index. . 0-9130 1-4073 . 0-9191 1-4732 . 0-9191 1-4740 . 0-9121 1-4742 . 0-9589 1-4791 . 0-9122 1-4686 . 0-9173 1-4717 Amer. Chem. Journ., x. 392-405. 41 626 foods: their composition and analysis. [§315. The following are the chief special tests for foreign oils : — Sesame Oil. — 20 cc. of the oil are mixed in a test-tube with 10 cc. of hydrochloric acid (specific gravity 1-19) which contains in solution 0-1 grm. of sugar; on shaking and allowing to stand a minute, should sesame oil be present, there will be a more or less crimson colour developed. The red colour in some olive oils adulterated with sesame appears in the oil rather than in the aqueous layer. According to V. Villavecchia and G. Fabris* sesam^ oil may be thus detected; 0"1 cc. of a 10 per cent, solution of furfurol is placed in a test-tube and 10 cc. of the oil to be tested added, followed by 10 cc. of HCl, specific gravity 1-19. The tube is shaken for half a minute ; a red colour denotes sesame oiL Sesame, if, present, will also alter the " constants" (see Table Llla., p. 624), its presence tending to raise the specific gravity, the iodine number, and the thermal value ; but a small per- centage of sesame will only afi^ect these slightly. Arachis Oil. — Arachis oil is so similar to olive oil m its general reactions that the only way in which it can be demon- strated with certainty is the isolation and estimation of arachidic acid. It is true that olive oil also contains arachidic acid, but in so small a quantity as not to be estimable from such quan- tities as 10 grms. of oil; whereas it is in about the proportion of 5 per cent, in arachis oil. To isolate arachidic acid the method of Renard is in use. 10 grms. of the oil are saponified, the soap decomposed by hydrochloric acid, the fatty acids dis- solved in 90 per cent, alcohol, and precipitated by lead acetate. (Lewkowitsch shortens the process by neutralising the soap by acetic acid, and then precipitating direct with lead acetate.) The lead salts are extracted with ethei-, which does not dissolve lead palmitate and lead arachidate. These latter are warmed with hydrochloric acid, the fatty acids allowed to solidify, and separated from lead chloride. The solid fatty acids are dissolved in 50 cc. of hot 90 per cent, alcohol. If arachidic acid be present, crystals of the acid are formed on cooling ; they have a definite form, and melting-point of 71° to 72° ; they may also be identified by dissolving known arachidic acid in a similar quan- tity in hot alcohol, and comparing with them the melting-point and microscopical characters of the crystals extracted from the oil. The crystals should be weighed, and the weight corrected by adding a number representing the arachidic acid still held in solution. 100 cc. of 90 per cent, alcohol dissolve 22 mgrms. at 15° C. and 45 mgrms. at 20° C. The quantity of arachis oil * Zeit. f. angexoandte Chem., 1893, 505, 50G. § 315.] COTTON-SEED OIL. 627 present is found approximatively by multiplying the -weight of the crystals by 20, calculating that arachis oil contains about 5 per cent. De Negri and Fabris have examined different mixtures of olive and arachis oils, and have obtained the following results (the arachis oil used containing apparently 4-78 per cent, of arachidic acid) : — TABLE LI. Sample containing Arachidic Acid found. Arachis Oil. GHve Oil. Arachis Oil. Weight of separated Crystals. Crystals in Solution. Total. 70 SO 85 90 30 20 15 10 Grm. 0-107 0-0605 0-0385 0-0200 Grm. 0-0315 0-0315 00315 00315 Grm. 0-13S5 0-0920 0-070 0-0515 Per cent. 29-08 20-24 14-00 10-30 In two other experiments with 10 per cent, arachis oil and 90 per cent, olive oil only unweighable crystals were obtained, and in a third a quantity, equal to 9-54 per cent, arachis oil; hence, when operating on 10 grms., the limit of detection appears to be 10 per cent. Should, therefore, qualitative evi- dence be obtained of small quantities of arachidic acid, from 25 to 50 grms. of the oil must be saponified to get quantitative results. Cotton-seed Oil. — This is detected by the alteration it pro- duces in the specific gravity, by the higher iodine number, by the higher melting-point of the fatty acids, and by Bechi's nitrate of silver test. For Bechi's test two solutions are required — viz., an alcoholic solution of silver nitrate, AgNOg, 1 grm. ; alcohol (98 per cent, by volume), 200-0 cc. ; ether, 40 cc. ; nitric acid, 0-1 grm., and a solution of colza oil, 15 parts, in 85 parts of amy lie alcohol. To apply the test, 10 cc. of the oil to be examined are mixed with 1 cc. of the silver nitrate solution, and then from 8 to 10 cc. of amylic alcohol solution of colza oil are added ; the mixture is shaken up, and heated in a water-bath for five or ten minutes. If cotton-seed oil is present, there is produced a brownish colour or turbidity of varying grade from light maroon to black. The unsaponifiable matter in olive oil is cholesterol ; in other oils, such as cotton-seed oil, phytosterol ; hence Salkowski's pro- 628 FOODS : their composition and analysis. [§ 315. cess, already describe in the article on Lard, is applicable to the detection of foreign oils, especially cotton-seed oil in olive oil. Rape Oil. — This may be detected by the character of the *' constants," especially by the melting and solidifying points of the fatty acids, the lower saponification and the higher iodine numbers. Fojyjnj-seed oil is chiefly indicated by the higher specific gravity and the high iodine number. It may be finally stated that experience has shown that, save a few special tests (such as Bechi's for cotton oil and the colour test for sesame), reliance must be mainly placed on a careful determination of the specilic gravity, of the melting and solidi- fying points of the fatty acids, the refraction, the iodine number, and the saponification values for the detection of adulterations of olive oil. BIBLIOGRAPHY. Benedikt, E-. — Chemical analysis of oils, fats, and waxes, translated and enlarged by Dr. J. Lewkowitsch. London, 1S95. Negri, G. de, and Fabris.— Gli Olii. Roma, 1893. PART IX.-EXAM I NATION AND ANALYSIS OF WATER. § 316. Pure water exists neither in nature nor even in the laboratory of the chemist, save on those rare occasions when, with immense expenditure of time and labour, water is purified either by repeated distillation over permanganates in a vacuum, or made synthetically. Nevertheless, however difficult it is to obtain even an ounce of water which shall consist of 1 part of hydrogen and 8 parts of oxygen by weight, and no other admixture, it may yet be very easily obtained sufficiently pure to wan-ant the epithet "pure" water — i.e., containing impurities only to be detected by reagents of great sensibility, or, what amounts to the same thing, by operating on a large quantity of water. In the analysis of water, therefore, it need scarcely be added that it is not the water per se which the chemist really analyses ; but his researches are directed with the object of unveiling and determining the nature, and where possible the amount, of whatever may be present, foreign to water, whether in suspension or solution, whether of mineral origin or as one of the myriad forms of " life." The experimental and analytical methods in use mainly fall under the following divisions — I. Examination by the Senses. — Smell, Sense of Taste, and General Appearance. II. Physical Examination, III. Chemical Methods. TV. Biological. — Embracing A, microscopical appeai-ances ; B, cultivation of fungi and dormant germs ; C, experiments on animals and human beings; D, expei-iments on fish. I. Examination by the Senses. § 317. Water that is evidently turbid, that possesses an odour and an unpleasant taste, requires no analytical processes to condemn it entirely ; such a water is unsuitable for drinking purposes. A water that even possesses any one of the enumerated bad qualities will, as a rule, be found to hold in solution sufficient impurities to make it decidedly objectionable. Most drinking- waters when looked at, or tasted, or smelt, without special precautions, have neither colour nor odour ; on the other hand, all water, if viewed through a sufficiently deep stratum, possesses colour. Colour. — To ascertain the colour of water, it is usual for analysts 630 foods: their composition and analysis. [§317. to be provided with a colourless glass tube, at least 2 feet in length, having the ends closed with plate glass, and a small opening in the side of the tube through which to pour the water. The purest waters have the slightest tinge of blue ; the next in order of purity have a just distinguishal)le shade of green. Decided green tints, London fog hues, amber yellow, and brown tints are those possessed by waters tinged with peat, containing suspended matters, of second class composition, or those of considerable impurity.* A far more scientific method is to fix two right-angled prisms, the reflecting surface of each being in opposite directions, in front of the divided slit, and then to fill a 400-cc. tube, such as is used for saccharimetry, with the water under examination, and a second 400-cc. tube with distilled water, and by means of two sources of light of equal luminosity, examine by the methods described (p. 84, et seq.) the different parts of the spectrum, and tabulate out the absorption of the water as com- pared with that of distilled water. Smell. — Haifa litre of the water or more is warmed in a large corked or stoppered flask to 38° [100° F.] ; a long glass tube of three-quarters of an inch in diameter is now inserted, and the water sucked up once or twice so as to wet the side of the tube thoroughly; then, without taking the tube out of the flask, one nostril is applied to the orifice of the tube, the other closed by the finger, and deep inspirations or "snifts" taken. Another simpler plan is to warm a quantity of the water, with- out removing the stopj^er, up to the temperature given, then shake, remove the stopper, and smell; a putrid odour denotes decomposing animal or vegetable matter. If the sample is much polluted by fresh sewage, a urinous odour is not unfrequently distinct. But, again, it may be specially noted that water quite unfit to drink may have no odour, hence the usefulness of the test is limited. A posi- tive smell teaches volumes — a negative result is of little value. Taste. — A few waters, and a few only, have a decided taste. It is scarcely to be recommended that analysts should taste samples derived from fever-stricken localities ; laut, on the other hand, when there is no suspicion of the samples having been the * Messrs. Crookes, Odling, and Tidy, in their report on the London water- supply for 1881, describe an ingenious "colour meter," consisting of two hollow wedges filled, one with a brown and the other with a blue solution. Any desired combination of green and blue may be made by sliding the wedges across each other in front of a circular aperture in a sheet of metal, and thus imitating the tint of watei under examination ; each prism is graduated from 1 to 50, the figures representing the thickness in milU- metres at that particular part of the prism. § 318.] WATER. 631 cause of any illness, the palate may detect some not unimportant peculiarity. II. Physical Examination. § 318. The physical examination is mainly microscopical. The spectroscope is also applicable, for Dr. W. Russell and W. Laplace* have observed that a column of pure water C feet in length shows a distinct single absorption-band ; and hence it is probable that, at all events, waters containing desmids and green vegetable cells generally would show particular spectra, but this has n.ot yet been worked out ; it will be more convenient for our purpose to consider the microscopical appearances later. (See p. 672.) III. Chemical Methods. A. — Preliminary Qualitative Chemical Exatnination. The qualitative examination of drinking water is not of much value, save when applied to a water considerably polluted. It may be restricted to the direct addition of Nessler solution, when a water containing a considerable amount of free ammonia will give an amber colour or even a precipitate, and to the testing for nitrites, nitrates, and metals. Detection of Nitrites. — The best tests for nitrites are— (1.) The meta-plienylenediamine test; (2.) Meldola's test; (3.) The naph- thylamine test. (4.) The zinc iodide starch test. (1.) The Meta-'phenylenediamine Test; the solution is made in the manner described in the Appendix. A cc. of the solution added to 50 cc. of water acidified with sulphuric acid, strikes a pale straw-yellow to a deep orange-red, according to the quantity of nitrite present. The limit of the reaction is, according to E. "Warington, 1 part of nitrite in ten millions of solution. (2.) MeldoloHs Test is a solution of para-amido-benzene-azo- dimethyl-aniline in water, acidified with hydrochloric acid, strength about -02 per cent. The reagent is added to the water to be tested, and the whole is acidified with sulphuric acid, warmed from ten to fifteen minutes on the water-bath and then alkalised by ammonia. If no nitrites are present, the tint is first a pale citron-yellow, changing on the addition of the acid to a rose-pink, and reconverted by ammonia or alkalies to citron-yellow. If, however, nitrites are present, the acid-liquid becomes of a salmon colour, and the final tint on the addition of ammonia is that of a sap green with small quantities of nitrate, and of a sapphire blue with larger quantities; on acidifying the liquid it changes back to a salmon colour. The tints are nob * Journal of the Chemical Society. April, 1881. 632 foods: their composition and analysis. [§318. permanent, but soon fade. The reaction, according to Warington, succeeds in a dilution of 1 part in one hundred millions. (3.) The Naj)hthylamine Test. — The water is first treated with sulphanilic acid, then acidified, and a solution of hydrochloride or sulphate of naphthylamine added. A minute trace of nitrite strikes a pale pink ; but, if much nitrite be present, a deep ruby- colour is produced, and the solution becomes turbid from the precipitation of colouring-matter. Mr. Warington found the reaction distinct with a dilution of 1 part of nitrite in five hundred millions of watei-, but in these great dilutions, the reaction requires to go on for an hour or two before the colour is developed. (4.) The Zinc Iodide Starch Test— To 100 cc. of the water are added 2 cc. of strong sulphuric acid, and a little zinc iodide and starch solution (see Appendix) ; in the presence of nitrites a blue colour appears. Detection of Nitrates. — Since no natural water is absolutely free from nitrate, the quantitative estimation of nitrates is alone of importance. Most tests are common to nitrites and nitrates. Nitrates, in the absence of nitrites, can be readily tested for by acidifying, adding a little zinc, and then testing with the zinc iodide starch test. For the brucine and other tests, see pp. 636, 637. Detection of Metals. — The metals most frequent in ordinary waters are lead or copper, and the most sensitive test for these is to add either ammon. sulphide or sulphuretted hydrogen water ; a dark tint or precipitate denotes either lead or copper, or both ; by adding potassic cyanide solution, if the dark hue be due to copper sulphide only, the solution clears ; if to lead sulphide only, it remains dark; if to a mixture, it partially clears. To confirm a copper reaction, test with potassic ferro- cyanide ; this produces a brownish colour or precipitate, accord- ing to the quantity of copper present. A convenient reagent is the author's cochineal test. A solution of cochineal in spirit strikes with a neutral or alkaline solution containing dissolved lead or copper, a deep mauve-blue to a red with a faint blue tinge, according to the amount present. The test will indicate ^^^j- of a grain of lead per gallon in ordinary drinking-water, and by the aid of comparison lead or copper-free solutions, smaller quantities of these metals may be detected. F. P. Venable* has found 4-48 grains of zinc carbonate per gallon in a spring water. Heatonf — in a Welsh spring, after flowing through half a mile of pipe galvanised iron — found 641 * Chem. Neivs, vol. \\. 18. + Ih., xlix. 85. § 319.] WATER. 633 grains per gallon of ZnCOg. Dr. Frankland has recorded a case of poisoning from a zinc-polluted well-water. Potassic ferrocyanide, added to the filtered and acidulated water containing zinc, gives either a light white cloud or heavy precipitate, according to the amount present. B. — Quantitative Analysis. § 319. A complete examination by chemical processes emhi-aces the following determinations : — 1. Total solid residue. 2. Estimation of the halogens, chlorine, and occasionally iodine, and in a few cases bromine. 3. Phosphates. 4. Nitrates and Nitrites. 5. Estimation of dissolved oxygen. 6. Sulphates. 7. Oxygen consumed in the Forchammer process. 8. Free and albuminoid ammonia. 9. Hardness. 10. Alkalinity, 11. Organic Analysis — Estimation of organic carbon and nitrogen. 12. Mineral analysis of water. The ordinary analyses, sufficient in most cases to pronounce an opinion as to the fitness of a water for drinking purposes, embrace only 1, 2, 3, 4, 7, 8, and 9. 1. Total Solid Residue.— Bj the totdil solid residue of a water is meant the substances in solution, as determined by drying up a measured portion, and weighing the dried residue ; if the water contain suspended matters, it should first be filtered, and a portion of the clear filtered liquid taken. The amount suitable for this determination depends upon the chai-acters of the water. The soft Devon waters yield a veiy insignificant residue from 100 cc, and to obtain trustworthy results, at least a quarter of a litre is required ; while, on the other hand, with calcareous waters, good results may be always obtained from 100 cc. With waters the characters of which are unknown, it will be best to operate on a quarter of a litre, or (if working with English measures) one- twentieth of a gallon. The water may be placed in a platinum dish, and evaporated down to a small quantity over a ring burner, taking care that the liquid in no case boils or even simmers; the last drops are driven off" on the water-bath. It is recommended by the Society of Analysts to heat the residue up to 104:°-4(220°F.) in the air-bath, and then to cool under a desiccator ; but with waters of unknown composition, it will be best to weigh the €34 FOODS : their composition and analysis. [§ 319. residue, wliicL lias not been exposed to a greater heat than 100°, for it is always open to the chemist to expose the residue thus obtained to liigher temperatures. The examination of the solid matters by the eye will often not unfrequently reveal much. Iron gives a coppery lustre to the dish, manganese a green to the ash, and very pure waters leave a residue almost white. The dish with its contents is next heated to a low redness, by the aid of a good Bunsen's burner, furnished with a rose, and then cooled and weighed. ISIote should be taken of any blackening or scintillation. Tlie loss of weight is returned as loss on igni- tion, and this hnal residue is dissolved in the manner to be described, and used for the qualitative determination of the phosjihates. 2. Estimation of the Halogens. — The estimation of chlorine is an essential part of the ordinary scheme of water analysis ; that of iodine is rarely (perhaps too rarely) performed, while so few waters contain an estimable amount of bromine, that it need not be here described. Chlorine. — Chlorine exists in ordinary waters in the form of sodic chloride, and occasionally a small portion of the total chlorine is combined with potassium. It may be estimated volu metrically by a standard solution of silver nitrate (See Ap- 2)endix), using as an indicator neutral potassic chromate. Nitrate of silver in presence of potassic chromate and alkaline chlorides (when the solution is neutral) first uses up or decomposes all the chlorides, and then attacks the chromates. Chloride of silver being white, and chromate of silver being red, the formation of silver chromate is indicated immediately by a red colour. At least 100 cc. of ordinary water (or, if grains are worked with, 140 grains) are to be taken for the determination of chlorine. With much-polluted waters, with those near the seashore or other pkxces in which the ground is impregnated with salt, such a quantity may be inconvenient, and it will be necessary then to dilute with distilled watei", taking of the diluted liquid a known quantity. In any case, the water is put into either a white porcelain dish or a beaker standing on a white slab. 1 cc. of the chromate solution (or 15 grains) is added to the water, and the standard solution run in from a graduated burette or pipette. The exact termination of the process is best observed through a glass cell, in which a little pale chromate solution has been placed. Since the eye, looking thus through yellow light, is very sensitive to the red rays, it may be necessary— especially where great accuracy is required — to i-epeat the determination in the following way : — The water from which the red colour of the silver chromate cannot be dischai'ged by stirring, is rendered again whitish-yellow, by the cautious addition of a very dilute solution of common salt. A fresh portion of water is titrated in a fresh dish or beaker, side by side with the former ; in this way the first permanent ditference of colour §319.] WATER, 635 can be observed. The results maj- be expressed in chlorine as chlorides, or it may be returned as common salt ; for the latter purpose multiply the clilorine by the factor 1-648, or more exactly by 1-64788. The following short table may facilitate calculation : — Chlorine. Sodium Cliloride. Chlorine Sodium Chloride. 1- 1-648 2- 3-296 3- 4-944 4- 6-592 5- 8-240 9-888 7- 11-536 8- 13-184 9- 14 832 10- 16-4S0 In a general laboratory it may be more convenient to estimate chlorine by Volhard's method,* because the solutions are in that case applicable to the determinations of the halogens in acid solutions, in which they have hitherto been estimated by weight. For ordinary purposes the solution of silver nitrate should be decinormal (that is, 17 grms. per litre), but for water analysis the strength given in the Appendix (p. 693) is most convenient. Besides the silver solution, are required — (1.) A solution of ammonium sul- phocyanide; (2.) Strong nitric acid -which has been boiled; (3.) A saturated solution of ferric alum. The sulphocyanide solution is diluted so that 10 cc. shall exactly equal 10 cc. of the silver. This is easily effected as follows:— 10 cc. of the silver solution, acidified with nitric acid, are run into a beaker, and a drop of the iron-alum solution added; the sulphocyanide solution is now run in from a burette until a single drop gives a red colour; the number of cc.'s used is noted, and the sulphocyanide diluted accordingly. The titration of the chlorides in water is done on similar principles, an excess of silver solution is added to 100 cc. of water, the mixture well shaken, and a few drops of alum solution added, and then the excess of silver solution determined by carefully running in the sulphocyanide until the red colour denotes the end of the reaction. Iodine. — M. Chatinf has upheld the theorj', that goitre is caused by waters insufficiently iodised— a proposition which cannot be considered proved. However, although M. Chatin has failed to convince the scientific world of the truth of his theory, he has done good service in showing how easy the detection and estimation of iodine in water really is, and in demon- strating the fact that most waters contain it in appreciable quantity. The process which M. Chatin used in his researches was : To evaporate one or two litres of the water to dryness with pure potassic carbonate, to calcine very moderately this dry residue, and then to extract with strong alcohol of 94 per cent. This alcoholic solution is again evaporated to drjmess, and moderately calcined ; the last residue is dissolved in a very little water, and will show all the reactions of potassic iodide. It is colorimetrically estimated by palladium, A solution of chloride of palladium gives a distinct colour with an infinitesimal quantity of iodiue ; hence it is only necessary to have a standard solution of potassic iodide, containing say 1 milligramme in 100 cc, and to estimate it precisely on the same principles as detailed [pout) for ammonia. Mr. Marchand,J pursuing the same line of researches, has preferred to precipitate from ten to twenty litres with nitrate of silver, collect the precijntate which may contain the chloride, iodide and bromide of silver, and dissolve it in sodic hyposulphite. The silver is now thrown out of this solution by sulphuretted hydrogen, and the solution, when freed from * " On the Estimation of the Halogens," by A. Percy Smith. Analyst, Jan., 1886. Liebig's, Annalen., cxc. 24. + Compt. Rend., t. xxxv., xxxix. J lb., xxxv. 636 foods: their composition and analysis. [§ 319. silver sulphide, evaporated to dryness with a little hydropotassic carbonate. In this way he obtains the chloride, bromide, and iodide of potassium. When the residue is perfectly dry, it is extracted with strong alcohol of 85 per cent. ; the alcoholic liquid is evaporated to dryness at a temperature not exceeding 75°. This last residue is again taken up by alcohol, and treated similarly to the potassic iodide obtained by Chatin's methed. 3. Phosphates. — The residue after ignition is treated with a very little nitric acid, and evaporated to dryness ; this treatment renders the silica insoluble. It is now again dissolved in a few drops of nitric acid, some water added, and filtered through an exhausted filter. If the filtrate is more than 5 cc, it should be concentrated to a smaller bulk, and its own volume of the molybdic solution added. The solution thus treated, and gently warmed, gives a more or less deep colour or a decided precipitate, according to the amount of i)hosphoric acid present. It may be estimated colorimetrically by a known solution of sodic phosphate, but this with no great accuracy. To make a gravimetric estimation of phosphates, save in polluted waters, may require several litres, and will seldom repay the trouble. Hence phos- phates may be returned in a qualitative manner as " t7-aces," with a feeble colour ; " decided evidence " with a dax'ker colour, and as "estimable amount" if there should be a precipitate. The Analyst Committee have adopted " traces," " Ii£avy traces," and " very heavy traces," as expressing three degrees of phosphate contamination. Such phrases are convenient, though somewhat paradoxical, and the avithor therefore prefers the more logical form of expression given above. 4. Ustirnation of Nitrates and Nitrites. — The several methods in use for the estimation of nitrates and nitrites may be arranged under the following heads : — (1.) Colorimetric methods of estimation. (2.) Estimation of nitrates by conversion of the nitrate into ammonia. (3.) Estimation of nitrates by decomposing the nitrate into niti-ic oxide, and measuring the gas. (4.) Indirect estimation by means of indigo. (5.) Estimation by the deficiency of hydrogen evolved from the action of sulphuric acid on iron powder. (1.) Colorimetric Methods — (a.) Tlic Brucine Method.* — 1 grm. of brucine is dissolved in 100 cc. of alcohol ; 10 cc. of the water are evaporated to dryness ; and from 0-5 to 2 cc. of the brucine solution added to the residue and six drops of a saturated solu- tion of oxalic acid. The colour must be a bright red; if it be brown, an insufiicient quantity of brucine has been added ; if pink, * See a paper by J. West Knight, Analyst, 18S1, 56-58. § 319.] 63< too large a quantity has been added. So it is best (in waters of unknown nitrate content) to put three or four separate 10 ce. in porcelain dishes and to evaporate them down with various quantities of the brucine solution (such as 0'5 cc. ; I'O cc. ; 1*5 CO. ; and 2 cc), and to make any quantitative determination on that which has the proper red colour. The colour is imitated by a standard test made by evaporating 10 cc. of a solution containing 72*1 mgrms. of potassic nitrate per 100 cc, with 3 cc. of brucine solution and six drops of the oxalic acid solution, and diluting up to 100. This test solution is diluted, if necessary. The original water treated in the same manner may, after evaporating and diluting, require filtration. G. Lunge and A. Lwotf* do not evaporate to dryness, but apply the test to the water direct ; they acidify with sulphuric acid, and the contents are heated to 70° to SO" until the liquid assumes a permanent greenish - yellow ; then this colour is imitated by a standard solution. They find that nitrites do not interfere. (b.) The Biphenylamine Method. — A solution of diphenylamine sulphate strongly acidified with sulphuric acid strikes a blue colour with nitrates. This colour can be imitated by a standard solution of nitrate, and the nitrates be thus estimated on colori- metric principles. This test is also well suited for quantitative spectroscopy. Hans Settegiirtf has worked out the absorption factors of diphenylamine as applied to small quantities of nitrates. Re- ducing his notation to wave-lengths, the following are his chief results : — TABLE LII. Wave-lengths. Absorption Coefficients. 0-120 to 0-500 Light Strength. 0-500 to 0-780 Light Strength. 616-6 to 543-2 543-2 to 532 532 to 521-6 521-6 to 513-6 513-6 to 505-7 •000002822 •000003162 •0(J0003606 •000003758 •000004226 •000002913 -0000032G6 •000003687 •000003811 •000004290 The values are correct for solutions containing nitric acid * Zeit. angew. Chem., 1894, 345-350. t Beitrage zur quautitativen Spectral-analyse. Chemie. Bd., vii., 1879. Annalen. der Physik. u. 638 foods: their composition and analysis. [§319. from 4-0 to O'l per 10,000. The phenylamine must be in 0*1 per cent, solution, and, of course, the sulphuric acid must be free from nitric or nitrous acids. (c.) Tlie Carbazol Test. — Samuel C. Hooker* has proposed cai-bazol as a test for nitrates ; it gives a green colour with oxidising agents. The objections to its use are that chlorine and iron have to be first removed, and that the presence of much organic matter makes the results too low. Assuming that the water has been freed from chlorine by silver sulphate and contains no iron, then 2 cc. of a sulphuric acid solution of carbazol are added, and the resulting green colour imitated by means of a solution of potassic nitrate. {d.) Phenol and Jiesorcinol. — D. Lindof has experimented with phenol in alcohol as well as with resorcinol. A 200,000th solution of N^O^ yields a faint pink band, green below, with a 10 per cent, solution of phenol in weak alcohol ; if 0-5 cc. of the solution be mixed with a single drop of the test in a test-tube, and then 2 cc. of sulphuric acid run down the tube so as to form bands of colour, N^Og, in the proportion of one 2,000th, gives very intense green and red bands. Phenol is also a good test for nitrates in the presence of free hydrochloric acid. liesorcinol, in 5 per cent, solution, is a good test for nitrites, so small a quantity as 1 in 500,000 giving (after standing four hours) a pink colour if acidified with sulphuric acid. One in 10,000 gives a pink colour at once. Iodides, bromides, and very large quantities of chlorides interfere with these tests, and must be removed by silver sulphate. Both the above are adapted for quantitative colour estima- tion. (2.) Estimation as Ammonia. — The most convenient method of obtaining the nitrogen of nitrites and nitrates in the form of ammonia is decidedly by the aid of the " copper-zinc couplt ." This method was first proposed by Gladstone and Tribe, and afterwards worked out in detail by Mr. M. Whitley Williams. J It appears that the copper-zinc couple decomposes nitrates first into nitrites, and then the nitrites into ammonia; nitrites are i:)resent to the last, and when all the nitrites have disappeared, it is certain the conversion into ammonia is complete. A low temperature, alkalies, alkaline earths and their carbonates, retard the reaction, while carbon dioxide, all acids, mineral acids, oxalic, phosphoric and common salt, as well as elevation of temperature, * Analyst, Sept., 1SS9. + Chem. News, Iviii. 1-3, 15-17, 28-29. X Journal oj Chtmxcal Society, Match, 1881. § 319.] WATER. 639 increase the reaction. In practice, a temperature of 24° is recommended as easily attainable. Manufacture of the Cojyper-Zinc Couple. — Pieces of clean zinc- foil, about 3 inches by 2 inches, are immersed in a 3 per cent, solution of cupric sulphate ; the zinc rapidly becomes coated with metallic copper. When a sufficient coating is obtained, the solution is poured off, and the couple well washed with water, finally drained, and the water for analysis poured on to the couple. It is best to do these processes in one and the same stoppered bottle. The water may nearly fill the bottle, and the stopper may be inserted, for there will be no gas evolved until the nitrates are entirely decomposed. The water thus treated is put in a warm place, and if the action is allowed to go on all night, the ammonia will be ready for estimation in the morning. The qviantity to be taken for the estimation of nitrates according to this plan, may be a quarter of a litre, or, if English measures are used, say 5 ounces. To very hard waters the addition of a little oxalic acid is recommended. In any case whei-e there is doubt whether the conversion into ammonia is complete, Griess's test, to be mentioned further on, should be used ; and if there is evidence of nitrous acid, the water must be left for a longer period. If the water possesses colour interfering with the Nessler agent, or matters precipitated by Nessler, it must be distilled in the or-dinary way, and the ammonia estimated in a fractional part of the much-diluted distillate. In most cases this is unnecessary, and by taking a measured quantity of the water and diluting it considerably, a fairly correct colorimetric estimation can be made by the direct addition of the Nessler reagent to the water thus diluted. It will be necessary to subtract from the amount of ammonia found, that which has been determined to exist in the water as ammonia. The ammonia derived from nitrates and nitrites must be expressed either as nitrogen or as nitric acid. The Aluminium Process. — The metal aluminium, when acted on by a caustic alkaline solution, decomposes nitrates into ammonia. A solution of soda of about 10 per cent, is prepared perfectly free from nitrates, by dissolving bit by bit metallic sodium in water. Any convenient quantity (such, e.g., as 100 cc, or 2,000 grains) of the water is placed in a suitable retort, which is fitted in an air-tight manner to a condenser, terminating in a flask as in the arrangement figured at page 476. An eqvial quantity of the soda solution is added, and the whole boiled until free from ammonia ; the retort is cooled, and the aluminium-foil dropped into the liquid, the whole is left over-night, and in the morning heat is applied to the retort, and the ammonia distilled over, and estimated in the usual way. 640 foods: their composition and analysis. [§319. ZJlsclis Method of Estimation of Nitric Acid by Redtiction to Ammonia* — Half a litre of the water is concentrated down to 15 CO. ; this is transferred to a flask of about 300 cc. capacity, 5 grms. of reduced iron {ferrtim redactum) added, and 1(3 cc. of diluted sulphuric acid, specific gravity 1'35. The liquid is heated to gentle boiling for some five or ten minutes, diluted with 100 cc. of distilled water, alkalised by from 20 to 25 cc. of soda lye (specific gravity 1*35), and distilled into 25 to 30 cc. of d. n. acid ; the diflerence in the litre of the d. n. acid, before and after the distillation, gives the data for measuring the nitric acid converted into ammonia. A correction by blank experiments should be made for impurities in the reagents. Example. — Half a litre of water was treated iu the way described ; and distilled, after being made alkaline, into .30 cc. of d. n. acid ; at the end of the operation, the acid used, instead of oO, 14 cc. of d. n. soda, 14 - 30= 16, which 16 of d. n. soda is exactly equivalent to 16 cc. of d. n. sulphuric acid which have been neutralised by ammonia. Since 1 cc. of d. n. acid is equal to 5'4 mgrms. of nitric anhydride (NoOs) ; 16 x 5-4 = 86 '4 mgrms. ; hence the litre contained twice that amount, or 172'8 mgrms. [12'1 grains per gallon]. (3. ) Estimation of Nitrates and Nitrites as Nitric Oxide — Crum Process. — Strong sulphuric acid acting on nitrates or nitrites in the presence of mercury, decomposes the nitrates or nitrites, and the whole of the nitrogen is evolved in the form of nitric oxide. Half a litre of the water is evaporated to dryness, the nitrates extracted by hot water, the hot-water extract evaporated down to 1 cc. , and the liquid transferred to the decomposition tube, which is a short tube about 3 inches long, constricted at one end, and furnished with a cup and stop- cock ; open at the other, and having a bore easily closed with the thumb. This tube is tilled with mercury, inverted, and clamped in a mercury trough with the cup uppermost ; it is now easy to transfer the solution of nitrates by pouring the solution into the cup, and cautiously opening the stopcock. The vessel in which the filtrate has been concentrated is then rinsed into the cup with pure strong sulphuric acid, and ultimately one and a half times the volume of the concentrated nitrate solution of strong sul- phuric acid is worked into the tube by carefully opening the stopcock. No air must be allowed to gain admittance. Should gas be immediately evolved, it is carbonic dioxide, and must be got rid of, for nitric oxide is not at once evolved. On the mixture of sulphuric acid and nitrate having been trans- ferred into the tube, the lower end is closed by the thumb, and the tube shaken so as to mix up the acid and the mercury, when the gas in a short time begins to come off, and considerable pressure may have to be exertecL When the reaction is com])lete, the contents are transferred to any gas apparatus and measured, t Every two volumes of nitric oxide equals one volume of nitrogen. The weiglit of the nitrogen is obtained from Table LVII. * Zeit. f. analijl. Chemie, xxx. 175 ; xxxi. 392. Tiemann Gartner's Handbuch. Braunschweig, 1895. t This is not necessarj', for the tube itself is now graduated, and the Bimplest method seems to be to plunge tiie tube into a vessel of water, cool it to the temperature of the water, adjust it so that the level of the water § 319.] WATER. 641 (4.) Indigo Process. — This process is based on the decolorisa- tioii of indigo when nitrates oi- nitrites are decomposed by strong sulphuric acid. When certain kinds of organic matter are present, the results are entirely without value. On the other hand, with careful working, the test is coi-rect with the great majority of waters, and as a means of rapidly determining the nitrates in unknown samples, with a view to their determination by other more exact processes, it is very useful. Fourgi-ms. of sublimed indigotinare digested for some hours with five times their weight of Nordhausen oil of vitriol ; the liquid is diluted with water, filtered, and brought to the volume of 2 litres.* A normal nitre solution is made by dissolving 1-011 grm. of pure potassic nitrate in one litre of water. From this solution, solutions of ^, ^, j^g-, ■^, and Jj-, normal are prepared. An assay is now made by mixing, say 20 CO. of the nitre solution with any amount of the indigo solution deemed sufiicient, in a wide-mouthed flask of 150 cc. capacity. Oil of vitriol is run into a test-tube, the volume being equal to the united volumes of the indigo and nitre. The con- tents of the test-tube are then suddenly tipped into the flask, and the flask transferred to a chloride of calcium bath maintained at 140°. If the solution of indigo is insuflicient, the liquid will be suddenly decolorised ; if it is too much, no bleaching will take place, the liquid still retaining its blue colour. In either case a fresh determination will be requisite, and by doubling or halving the amount of indigo for the next experiment, as the case may be, the operator will soon find the limits, and five or six experiments will standardise the solution. In every instance a quantity of sulphuric acid, equal to the united volumes of indigo and water, must be used ; the indigo solution should be diluted so as to be about equal to the nitre solution. As it is found that the quantity of indigo consumed is not precisely in proportion to the nitric acid present, but diminishes as the nitrate solution becomes more dilute, the further standardising of the indigo solution by the more dilute solutions of nitre already alluded to is necessary. The results may be thrown into a table as follows. (See Table LIII., next page.) The method of using the table is sufiiciently obvious to those who are accustomed to calculations of the kind; supposing, for example, 20 cc. of the water used up 6 "64 cc. of the indigo ; this is '5 cc. above the nearest number in the table, viz., 6-14. Now, taking inside and outside the tube is the same, and measure it direct. The small absorption of the nitric oxide by water, requires no correction. Of course, the value of the divisions on the tube must be ascertained with accuracy. • See Mr. Warington's excellent paper, Journ. Chemical Society, Sept., 1879, p. 579; also, Frankland's "Water Analysis," p. 31. 42 642 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 319. TABLE LIII. — Value of Indigo, in Nitrogen, for Different Strengths of Nitre Solution. Difference in Strength of Nitre Solutior Used. Indigo required. Difference between Amounts of Indigo. Nitrogen Corresponding to I cc. of Indigo. Difference between tlie Nitrogen Values. the Nitrogen Values for a Difference of 1 cc. in the Amount of Indigo. cc. cc. -i^ normal, 10-00 •000035000 s „ 8-71 1-29 •000035161 •000000161 ■000000125 l\ 7-43 1-28 •000035330 •000000169 ■000000132 ^ „ 6-14 1-29 •000035627 •000000298 ■000000231 ^ „ 4-86 1-28 ■000036008 •000000381 ■000000298 s 3-57 1-29 •000036764 •000000756 •000000586 s ,, 2 29 1-28 •000038209 •000001445 •000001129 ^ „ 1-00 1-29 •000043750 •000005541 ■000004295 tlie extreme right-hand column, the difference for the nitrogen values of 1 cc. will be found ; and as there is in this case a difference of only half that quantity, halving the number gives us -000000115 ; this number has to be subtracted from the unit value of nitrogen found in the first column, thus : — •000035627 - -000000115 = -000035512, which is the nitrogen-value of each cc. of the indigo. Hence, as we have supposed that 20 cc. of the water decolorised 6-64 cc. of indigo, the nitrogen as nitric acid in parts per 100,000 is 1-179, or in grains per gallon -82 grain. If the indigo estimation of nitric acid is only a preliminary step to a further and more exact determination by the Crum method, these refinements are not necessary. The indigo solution is standardised once for all by the normal solution of nitre, and if the nitrates are either very large or very small, an allowance is made. (5.) Ulsch's Method of Estimating Nitric Acid by Measuring the Dejiciency of Hydrogen evolved on Reduction. — One of Ulsch's methods has been already detailed. This method is in its principle an indirect one, for what is "measured is not the ammonia produced, but the deficiency in the evolution of hydrogen. A quarter of a litre of water is evaporated down to 15 cc. and, while still hot, filtered. The filter is washed with a little boiling water into a small measuring flask, the size depending § 319.] 643 on the amount of nitrate present, which can be approximately ascertained by a brucine or other colour test ; 50 cc. is the capacity for most waters — viz., those containing under 150 mgrms. of N2O5 per litre • larger flasks are used for larger quantities. The collective fluid with washings should not be more, for a 50 cc. flask, than 40 cc. The flask is cooled to the Fig. 72. temperature of the room, and sufiicient normal sulphuric acid added to make the contents equal to 4 normal when water is added to fill the flask to the mark on the neck. This fluid is the " testing fluid," of which 10 cc. are taken for the estimation of nitric acid. In a thin walled flask^ a (fig 72) the bulb of 644 foods: their composition and analysis. [§319. ■which is equal to about 30 cc. 3 grms. of iron powder [Ferrum pulveratum) are placed by means of a dry funnel. The flask is closed by a caoutchouc stopper containing three holes ; one of which carries the twice bent 3 mm. wide tube, h, the one end of which goes almost to the bottom of the flask, the other end is connected by means of a bit of rubber tubing to a short piece of glass tubing — the rubber is supplied with a screw clip. The middle hole of the caoutchouc stopper carries the funnel, c, which is provided with a stopcock ; the end of the stem of the funnel is bent slightly, so that fluids will run down the neck of the flask. The third hole carries a right-angled tube, connected by narrow rubber tubing, to the gas-measuring apparatus. This consists of a burette, f, divided into tenths of a cc, the upper end of which is provided by a — |-tube, e, and the lower end con- nected by rubber tubing to the spherical funnel, g. The first operation is to fill the whole system with hydrogen, water is poured into g, the stopcock at e opened, and the funnel raised until the whole of / is filled up to the stopcock. Into the funnel, c, are poured 12 cc. of dilute sulphuric acid (33 cc. strong sulphuric acid to a litre of water) and some 10 cc. of it allowed to flow on to the iron, by opening the stopcock ; an evolution of gas commences, and in about two minutes ceases. Before it actually ceases the stopcock, e, is closed and the clip at h. The funnel, g, is lowered as much as possible. Now the rest of the acid is allowed to flow into the flask, and the funnel washed twice with a 2 per cent, copper solution, which is also allowed to flow into the flask, care being taken not to aUow the admittance of air. The flask is now warmed in a beaker of water at 60^ C. for two minutes without shaking ; then the flask is shaken without interruption for another two minutes ; after which the flask is cooled as quickly as possible to the temperature of the air of the room. The funnel, g, is now raised, and the clip on h cautiously opened. The fluid contents of the flask, a, are under the pressure of the hydrogen forced through b, and may be received in a beaker. If the twice right- angled tube, h, is properly adjusted none of the iron will escape; the clip is closed while there is still a small column of liquid in the tube. After thus emptying the flask, its temperature is adjusted to that of the air, and the excess of gas is got rid of by carefully opening the stopcock at e ; while the water level of / and g is adjusted to the zero point of the burette,/. The apparatus is now ready for the purpose of ascertaining once for all how much hydrogen is developed by 10 cc. of i normal sulphuric acid acting on the iron in the absence of nitric acid. From a burette exactly 10 cc. of the acid are allowed to flow § 319.] WATER. 645 into the funnel attached to the small flask, a, while g is lowered to the level of the table ; then the acid is allowed to flow slowly into the flask. From a second burette 10 cc. of 2 per cent. copper sulphate are added to the funnel and transferred to the flask, always avoiding entrance of aii-. The flask is warmed to G0°, shaken, and finally cooled exactly as detailed pre- viously. The amount of gas, on adjusting the water level of g to the water level of f, is read, the temperature of the water in g being taken, and also the height of the barometer, and the volume of gas reduced to normal temperature and pressure — always subtracting '10 cc. from the known volume of the flask, that being occupied by the 20 cc. of liquid. The flask may now be emptied as before, and a second or third determination made. The mean being taken as the standard. To estimate nitric acid the process is precisely the same, save that 10 cc. of the testing fluid are taken instead of 10 cc. of i normal acid. In the presence of nitrates there will be a deficiency in hydrogen, from which the nitric acid can be calculated — the following reactions taking place : — 2KNO3 + H0SO4 = K2SO4 + 2NO3H 2NO3H + SH2 = 2NH3 + 6H.,0 2NH3 + Hi04 = (NH4).,S04 or 8 molecules of hydrogen are required for the reduction of one molecule of nitric acid, besides which, for the decomposition of an equal molecule of nitric acid from its combination with a base, and the combination of the ammonia formed with a farther molecule of sulphuric acid, 2 molecules of hydrogen are required. A molecule of saltpetre then will in respect of a normal hydrogen volume show a deficit of 10 molecules of hydrogen. A deficit of 1 cc. of hydrogen will, therefore, be equal to 0*90416 mgrm, KISfOg or 0-4834 mgrm. nitric anhydride (NoO^). An example of this method may here be given. 50 cc. of test fluid derived from a quarter of a litre of spring water were prepared according to the method described in the test. 10 cc. of the latter yielded as a mean of three estimations (subtracting the 20 cc. of fluid in the flask) 20'2 cc. hydrogen, the barometer being 741 mm. (temp. Ib^'S), and the corresponding tension 1.5"9 mm. This, reduced as follows: — 20-2 X (741 - 15-9) ,„ ._ ,, , ' - 18-05 cc. of hydrogen. (1 + 0-0030 X 18-5) X The normal hydrogen volume in the absence of a nitrate was 21 -02 cc; the hydrogen deficit is, therefore, 21-62 - 1805 = 3-57 cc. 646 foods: their composition and analysis. [§319. 3 '57 X 0"4833 = 1725 mgrm. of nitric acid in 50 ec. of water ; therefore a million parts (a litre) contain 0-00a725 X 1-000000 ^, _ ,,,^ , — = 34*0 mgnns. (JSI0O5). 50 The process is made inaccurate by the presence of iron or carbonates. The carbonates of the earths as well as any iron are, however, by evapora- ing to a small volume as described, fully separated. Should alkaline carbonates be present a little gypsum must be added to the water before evaporation. Small quantities of nitrites introduce no material error ; in the presence of much nitrite advantage is taken of the fact that the iron copper couple reduces nitrites in the cold. The test fluid is, therefore, freed from nitrites by digesting the test fluid with copper solution and iron in the cold. A correction may also be made by estimating the nitrites by a colorimetric process, for which the zinc- starch iodide seems suitable, as well as the other colour tests for nitrites already described.* (5.) Estimation of the Dissolved Oxygen in Water. — The amount of air dissolved in water is dependent on temperature and pressure, the amount of oxygen in the air of pure water having a mean value of 34'91 per cent. The following table gives the absorption coefficient of air in water ; and also the amount in cc. of oxygen that a litre of water will dissolve. The oxygen, in what may be called the water atmosphere, is diminished by the activity of micro-organisms ; hence its deter- mination is valuable. It is especially useful in the investigation of the polluted water of a river, particularly if determinations of oxygen are made on the spot. On the other hand, water sent from a distance to an analyst, and probably one or two days on the road, cannot have its dissolved oxygen directly determined with advantage. The writer, in such a case, pro- ceeds as follows : — Half a litre of the water is shaken up in a large Winchester quart, until saturated with air; the dissolved oxygen is determined in half of this — that is, ^ of a litre ; the other -J- of the litre is put on one side, a layer of xylene having been poured on the surface of the water to exclude air, and again titrated for oxygen at the end of forty-eight hours ; the difference will have a direct relation to the organic matter and micro-organisms in the water. This method may be called " the differential method of estimating oxygen in water." * Nitrites may be estimated by Piccini's method, which depends upon the fact that if a solution of nitrite is treated with urea, and thoroughly acidified with sulphuric acid and gently heated, all the nitrogen of the nitrite is evolved as gas. The acidification must take place in a vacuum, or, at all events, atmospheric air must be excluded, otherwise the reaction is not quantitative, some of the nitrite passing into nitrate. Nitrites may be also estimated by titration with potassic permanganate. § 319.] 647 TABLE Lllla. showing the Absorption Coefficient of Air dissolved IN Water and thi; co.'s of Oxygen which may be obtained from 1 Litre of Water (after Bunsen).* Temp. C°. Absorption Coefficient of Air dissolved in Water (cc. of Air dissolved in 1 cc. of Water). The Number of cc. of Oxygen which 1 Litre of Water will dissolve. 0-02471 8-63 1 0-02406 8-34 2 0-02345 8-19 3 0-02287 7-98 4 0-02237 7-80 5 0-02179 7-60 6 0-02128 7-43 7 o-oeo80 7-26 8 0-02034 7-10 9 0-01992 6-95 10 0-01953 6-81 11 0-01916 6-69 12 0-01882 6-57 13 0-01851 6-46 14 0-01822 6-36 15 001795 6-26 16 0-01771 6-18 17 0-01750 6-11 18 0-01732 6-05 19 001717 5-99 20 0-01704 5-95 * Gasometriache Methoden., 2 Aufl., S. 387. Winkler has also published values for the amount of oxygen dissolved in water, hi» numbers closely correspond to those of Bunsen's for temperatures 16° to 20°, but are more than 1 cc. higher for lower temperatures. 648 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 319. Preusse and Tiemann have arranged a convenient apparatus, by means of which the dissolved air is boiled out of water into soda lye, which has been completely deprived of its gases by boiling. The oxygen in the collected gas is determined either by exploding with hydrogen or by absorbing it with alkaline pyrogallate. Schutzenberger and Risler cause the dissolved oxygen to act on sodium indigo-white-disulphonate in excess, and change a portion into sodium indigo-blue-disulphonate, thus — Ci6HioNA(S03Na)„ + 0„ = CicH8NA(S03Na)o + H„0. The salt is changed back again by a titrated solution of sodium hydrosulphite, according to the equation — Ci6H8NA(S03Na)2 + H„--.CicHioNA(S03Na)o. The end of the reaction being sharply indicated by the change of colour from blue to yellow. Both the above processes and their various modifications demand more or less complicated apparatus, and are not adapted for estimations on the banks of streams or anywhere else, save in a laboratory ; the writer, therefore, prefers a modification of Winkler's method ; it has great accuracy, and may be performed almost anywhere. The essential fact on which Winkler's method is based is that manganese hydrate absorbs oxygen ; a third hydroxyl being added in the presence of oxygen and water, to the two hydroxyls in manganese hydrate — that is to say, Mn(0H)2 become Mn(0H)3. This last hydrate, by excess of HCl, is transformed into manganese trichloride, two molecules of which react with two molecules of potassium iodide, setting free two atoms of iodine ; so that for every IG parts of oxygen 254 parts of iodine are set free; the iodine set free is, therefore, titrated and con- verted by calculation into oxygen. The equations setting forth, the general reactions are as follows :— 2MnCl2 +4NanO = 4NaCl + 2Mn(OH)5 2Mn(OH)2 + + H„0 = 2Mn(OH)3 2Mn(0H)3 + 6HC1 =2MnCl3 + GH„0 2MnCl3 + 2KI = 2MnClo + 2KC1 + 21 The following solutions are required : — (1.) 10 grms. potassium iodide dissolved in 100 grms. of 33 per cent, pure soda solution. § 319.] WATER. 649 (2.) Solution of 80 per cent, manganous chloi'icle (free from iron).* (3.) Hydrochloric acid (specific gravity 1-16 to 1*18), (•i.) Starch solution. (5.) Solution of thiosulphate — the iodine value of which is known ; each cc. should be equivalent to a centinormal iodine solution — that is to say, 0-000127 I. The temperature of the water and the height of the barometer being first ascertained, 250 cc. of the water is covered with xylene to the depth of an inch, and 1 cc. of the alkaline iodide solution is allowed to flow in through a pipette, the end of the pipette being held below the xylene, next 1 cc. of the manganous chloride solution is allowed to flow in. The mixture is stirred with a glass rod until the contents are completely mixed and the flask put on one side for a little time. Then 3 to 5 cc. of hydrochloric acid are added ; the precipitate dissolves, and the iodine is set free ; this is titrated by the thiosulphate, using as an indicator starch. If the thiosulphate is equal to centi- normal iodine solution, each cc. equals 0-0000798 grm. oxygen, or 0-055825 cc. oxygen ; if the cc.'s of water used = Y, the cc. of thiosulphate = n, then the content of oxygen in a litre of water is obtained by the following calculation : — 0-05582.5 re x 1000 The number thus obtained must, of course, be reduced to standard pressure and temperature. A reference to the table on p. 647 will show whether (at the temperature of the experi- ment) the water is below the standard or not. A correction is usually necessary for most waters, especially those that contain nitrites, organic matter, and other impurities. For this purpose 1 cc. of the manganous chloride solution is mixed with half a litre of distilled water, alkalised with 1 cc, of a 33 per cent, soda solution, shaken, and the brown precipitate collected on a small filter. The precipitate is dissolved in hydrochloric acid, and the solution diluted to half a litre. Two separate portions (each 100 cc.) of this solution are taken, and mixed, the one with 100 cc. of distilled water, the other with the water to be tested. After a few minutes, a few crystals of potassium iodide are added to both mixtures, and the iodine * The solution may be freed from iron by boiling and precipitating with soda, the filtrate is acidified with HCl, evaporated to a syrup, and then crystallised. 650 food:.: their composition and analysis. [§ 319. sepai-ated from each estimated by titration. The difference of the values equals the amount to be added. In waters of this tind the titration of oxygen is best done by adding the man- ganese chloride solution and simply 33 per cent, soda lye, the potassium iodide being finally added in crystals ; the following is an example : — 250 cc. of the water of the River Exe just below Exeter, at 15° and 760 mm., treated in the method stated, set free iodine equal to 12 cc. of centiuormal thiosulphate. The correction was obtained by mixing 100 cc. of the manganous di- chloride solution with 100 cc. of distilled water, and by treating 100 cc. of the river water in the same way, adding to each potassium iodide, and titrating the iodine set free. The distilled water gave 7 cc, the river water 4 cc. ; the difference is, therefore, 3 cc. Hence, the content of oxygen at 15° of a litre of Exe water is (12 + 3) X 0-055825 x 1000 „ „. „ 250 = ^'^^ ""'■ ^ For most practical purposes it is not necessary to reduce the cc. of oxygen tlius found to normal temperature and pressure ; hence, in such a case as the above, it would suffice to report that the water contained 3-35 cc. oxygen at 15° C, or about half the normal quantity. (6.) Sulphates. — Any convenient quantity of the water, carefully mea- sured, is acidified with hydrochloric acid, and heated neai'ly to boiling ; while hot, some solution of chloride of barium is added, so as to be ui slight excess, and the solution kept near the boiling point for some time. The sulphate of barium is allowed to settle, collected on a filter, dried, ignited, and weighed : one part of baric sulphate equals -134335 of sul- phuric acid. An estimation of sulphates in water can also be made on colorimetric principles. To 100 cc. of water, barium chloride is added in slight excess and then the water is acidified by hydrochloric acid, the turbidity produced is now imitated by a dilute solution of sodic sulphate tested with the same reagents. It is best to observe the turbidity by looking through the colorimeter at a black porcelain plate, (7.) The Forchammer, Oxygen or Permanganate Process. — The principle of this process is the abstraction of oxygen by the organic elements of the water, and the estimation of the oxygen thus abstracted. "395 grm. of potassic jDermanganate is dis- solved in a litre of water, which gives a solution containing 1 mgrm. in every 10 cc, or, if working in grains and septems, 2 gnains of permanganate in 1,000 septems of water, equalling •01 grain of available oxygen in 20 septems. This is the standard solution. The determination is now usually made, as recommended by the Society of Analysts, in two stages, on two equal quantities of water, viz. — (1.) The amount of osygen absorbed in fifteen §319.] WATER. 651 minutes, and commonly clue to nitrites, or, at all events, sub- stances very readily oxidisable; and, (2.) the amount of oxygen absorbed in four hours. The time for this last determination used to be given as three hours, but the four-hours period is preferable ; and even then it is easy of proof that, if the water be allowed to stand, there still remain matters capable of being oxidised. The temjierature is an important factor, for numerous experi- ments have shown that the amount of oxygen consumed varies greatly at different temperatures. The Analysts' Society have adopted 2Q°-6 (80" Fahr.), and in order to ensure tiniformity this temperature is here recommended. It is, however, probable that better and more uniform results would be attained by boiling the water and permanganate for an hour. In some interesting experiments by Messrs. Wigner and Hai'land,* i-iver water, to which a known quantity of pure sugar had been added, was found to have absorbed more oxygen at the end of two hours, at 37°-7 (100° Fahr.), than during six hours at 15°-5 (G0° Fahr.), and almost as much as during six hours at 26'' -6 (80° Fahr.) Similarly, river water contaminated by a known quantity of urine used up equal quantities of oxygen when acted upon by permanganate for six hours at 26°-6 (80° Fahr.), as it did when the process was accomplished in two hours at 37°'7 (100° Fahr.) The actual operation is as follows : — Two stoppered flasks are taken, and a quarter of a litre of the water put in each [or 3,500 grains]. The bottles, with their contents, are immersed in an air-bath until the temperature rises to 26° -6 (80° Fahr.), then 10 cc. [or 100 grains] of dilute sulphuric acid [1 : 3] are added, and the same quantity of the standard permanganate. One of the bottles is taken out at the end of a quarter of an hour, and two or three drops of potassium iodide added to remove the pink colour. After thorough ad- mixture, there is run into it from a burette a solution of sodium hyposulphite, the value of which has been determined by titrating 10 cc. of the standard potassium permanganate in distilled water tintil the yellow colour is nearly destroyed ; then a few drops of starch water are added, and the hyposulphite added until the blue colour is just discharged. At the end of four hours the other bottle is removed and titrated in exactly the same way. Should the pink colour diminish very rapidly dui-ing the four hours, another measured quantity of perman- ganate mvist be added. If A be taken to express the amount of hyposulphite used for a blank experiment with pure distilled water, B the water * On the Action of Permanganate on Potable Waters at Different Tem- peratures. Analyst, March, 18S1, p. 39. G52 foods: their composition and analysis. [§319. under examination, and a the amount of available oxygen in the quantity of permanganate originally added : then, the oxygen consumed by the quantity of water operated on would be (A - B) g A or, in actual figures, 10 cc. of a permanganate solution, equivalent to -001 grm. of oxygen, were added to a quarter of a litre of distilled water, and to the same quantity of a sample under analysis. The distilled water used 40 cc. of hyposulphite, the water 15 cc. at the end of four hours. Then the oxygen con- sumed by the quarter litre was -000625, according to the equation 40- 15 X -001 ^.^^^^^^ 40 or per litre, -0025 [-175 grain per gallon]. Blair prefers to use the oxygen process at 100° C, and has published an elaborate series of experiments proving that per- manganate and sulphuric acid in the absence of organic matter or reducing agents can be boiled for two hours without change ; a serious objection is that a boiling permanganate of potash solution will decompose chlorides, setting chlorine free, thus — KsMiioOg + 8H2SO4 + lONaCl = K0SO4 + 2MnS04 + 5Na2S04 + 8H2O + 5^L ; but with waters containing up to 8 grains of chlorine per gallon, it appears that this reaction has absolutely no effect; with largt-r quantities of chlorine, a control may be run containing an equal amount of chlorine in the form of common salt. A great number of organic substances, when treated with permanganate at 100°, absorb the quantity of oxygen theoretically necessary to convert the carbon into carbon dioxide, and the hydrogen into water; 100 mgrms. of cane sugar in two hours absorbed 111-2 mgrms. O, as against 111-9; strychnine, brucine, morphine all gave the theoretical amounts ; but starch came out low, or 86-4 instead of 118-5. The writer considers that with regard to ordinary waters the oxygen process at 100° is by far the most reliable, and has, for the last four years, used it exclusively. He is also able to confirm Kruss, that the strength of a per- manganate solution may be estimated by the spectroscope by the use of the divided slit (see p. 86) ; those who have the necessary appliances may make use of this method in preference to titration. §319.J WATER. 653 The German chemists generally estimate oxygen consumed at 100°, but only boil for ten minutes. The method of Kubel is the oxidation by a centinormal solution of permanganate in ten minutes in acid solution : that of Schulze is, first, oxidation by alkaline permanganate ; then the solution is acidified by sul- phuric acid and again boiled ; in each case the boiling is for ten minutes. Probably the method of Schulze is the better of the two. In either case the oxygen consumed (or, in other words, the permanganate used) is estimated by oxalic acid as follows: — The permanganate strength is first ascertained by centinormal oxalic acid. The water which has been boiled with perman- ganate has its colour discharged by means of centinormal oxalic acid added in known volume and in slight excess, and then the solution of permanganate is dropped in until a weak red colour is permanent. The calculation is obvious, but it may be useful to give an example. A centinormal solution of permanganate exactly equivalent to a centinormal solution of oxalic acid was used, and 15 cc. of such a solution added to 100 cc. of water; after boiling for ten minutes and then cooling, 10 cc. of the oxalic acid solution was run in ; and then to the colourless fluid it was found that 5 cc. of the permanganate solution was necessary to just redden the colourless liquid. In all, the solution contained, therefore, 20 cc. of permanganate, of which 5 have been used ; 1 cc. of centinormal potassic permanganate is equal to 0-08 mgrm. of oxygen; hence the 100 cc. used 0'4 mgrm. of oxygen, equivalent to 4 parts per million. The Oxygen Consumed Applied to the Indirect Estimation of Volatile Organic Matter. — Preusse and Tiemann* have submitted various waters to distillation, and have estimated the amount of oxygen consumed in the distillate. They have come to the conclusion that the products of putrefaction may be, in this way, detected, and that the process is a valuable aid to the judgment of drinking waters. Those waters which are good, reducing but small quantities of permanganate ; those that are impure, large quantities. An example of one of their experiments may be given. Half a litre of water derived from the Panke, a dirty brook running through the north-west of Berlin, was distilled, 400 cc. of distillate being collected in four successive fractions and boiled, after Kubel's method, ten minutes as before described. Half a litre was first distilled from the neutral water : another half * Tiemann-Gartner's Randbuch der Untersuchitng u. Beurthdlung der Wasser. Braunschweig, 1895. 654 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 319. litre was acidified; a third was alkalised. In all three cases volatile organic matter was present. (a) Distillation of the Neutral Water. 1st. 100 cc. consumed 1 '28 mgrm. of oxygen. 2nd. „ „ 0-65 3rd. „ „ 0-39 4th. „ „ 0-27 (6) Distillation of the Acid Water. 1st. 100 cc. consumed 1 '14 mgrm. of oxygen. 2nd. „ ,, 0-69 3rd. ,, „ 0-45 4th. „ „ 0-40 (c) Distillation of the Alkalised Water. 1st. 100 cc. consumed 1"08 mgrm. of oxygen. 2nd. „ ,, 0-4S 3rd. „ „ 0-27 4th. „ „ 0-26 (8.) Ammoiiia, Free and Albuminoid, — The estimation of ammonia depends on the principle that it admits of ready dis- tillation when it exists in the water as ammonia, provided that the water is alkaline. Since, therefore, nearly every natural water is alkaline, distillation of water is alone sufficient to expel the ammonia. If a water, by testing with cochineal, is found to be acid, then it will be necessary to add a little recently ignited carbonate of soda (or, perhaps better, a little recently burnt magnesia), until au alkaline reaction is obtained. The apparatus required for the estimation of free and albuminoid ammonia, is — (1.) A good large stoppered retort, fitting into a full-sized Liebig's condenser, through which a constant stream of water is running. (2.) Measuring-flasks, either in septems or litres. § 319.] WATER. 655 (3.) Cylinders made of clear glass, " Nessler cylinders," or a colorimeter. (4.) One or two pipettes. (5.) Nessler reagent {see Appendix). (6.) Standard solution of ammonium chloride (see Appendix). (7.) Solution of alkaline permanganate (see Appendix). The water is first tested with a little of the Nessler reagent; if it shows any decided colour it may be necessary to distil a very small ])ortion, say a quarter of a litre, diluted with a sufficient quantity of pure, ammonia-free water. But if, on the other hand, there is no colour, or a doubtful one, a litre of the water should be distilled, or a fifth of a gallon. On distillation, 100 cc, or 1,400 grains, are collected in one of the glass cylinders, and 5 cc, or one-twentieth of its volume, of clear straw-coloured Nessler solution added. If there is any ammonia the distillate thus tested will be tinted or coloured, the colour varying from a very pale straw up to a dark amber. If the colour should be very deep, it is impossible to estimate the ammonia with even an approach to accuracy, unless the dark solution is very much diluted and made up to definite volumes, of which definite volumes fractional parts are taken. The next step is to estimate tlie ammonia by imitating the colour. This is done by running into some distilled water one, two, or more cubic centimetres of the standard ammonium chloi'ide solution, and adding exactly the same amount of Nessler solution as had been added to the distillate. The solution is now made up to precisely the same bulk as the dis- tillate, and the liquids, thus in equal columns, compared by looking down through them on to a glass plate or white porcelain, tile or slab. Accurate estimations may be made by the special colori- meters described on pp. 80-83, but the accuracy greatly depends upon the practice of the observer, and the sensitiveness of his eye for diff'erences of colour. There are many persons who, from some physical peculiarity of sight, can only distinguish a few shades, and even with the greatest care can make no very accurate colorimetric observation. By graduated Nessler glasses, having taps near the bottonx in order to run off" a portion, as well as by colorimeters, such as Mill's and other like contrivances, " Nesslerising " is much expedited and facilitated. Returning to the actual estimation of free ammonia, the water must be distilled in successive fractions, until no more free 656 FOODS : their composition and analysis. [§ 319. ammonia is detected in the distillate. This occurs generally when 150 cc. or 200 cc. (that is, one-fifth of the entire quantity) of the water taken lias come over, then the water is ammonia- free. The next step is to estimate in the same water the albuminoid ammonia. Albuminoid Ammonia. — When Mr. Wanklyn first published the albuminoid process it was very generally adopted, and it may be considered as yielding quickly certain data, assisting in the final verdict of an analyst. To the water about one-tenth of its original volume of the alkaline permanganate {see Appendix, p. 693) is added ; the water is again distilled ; successive fractions of the distillate are tested with Nessler, and the ammonia therein con- tained determined in the same way as in the free ammonia esti- mation. Here the analyst often has considerable difficulty, from the circumstance that evidently the alkaline permanganate often sets free certain compound ammonias, which strike a tint with the Xessler re-agent entirely difierent from that given by pure ammonia. In certain cases it may, indeed, be necessary to esti- mate the ammonia by titrating with a feeble and very dilute acid. The free ammonia is usually returned as ammonia ; the albuminoid sliould properly be returned as " nitrogen as albuminoid ammonia." It is scarcely necessary to remind the operator that all retorts, conden.sei's, &c., used for these estimations must be ammonia-free, and that ammonia from any analytical operation mvist not be allowed to contaminate the laboratory atmosphere. The most ready way to render it cei'tain that there is no ammonia in the condenser is to acidify a little water with sulphuric acid, and then distil until the distillate is ammonia-free. (9.) Hardness. — A. Before Boiling. — In a corked or stoppered bottle 1000 grains, or 100 cc, of the water to be tested are placed. The standard soap solution {see Appendix,^). 693) is run in, 10 grains, or 1 cc, at a time, and after each addition the cork or stopper is replaced, and the bottle shaken violently, and observed as to whether a permanent lather forms or not. If not, then another measured quantity is run in, and so on until the desired efi"ect is produced. Waters containing but little magnesia give a good lather, and the reaction is fairly sharp. With magnesian waters the reaction is slow, and not so easy to observe. When a lather has been obtained, it is well to repeat the experiment, and in this second assay to rim in within half a division the whole of the amount of soap solution thought to be necessary; then a further jjortion of the soap solution is run in very gradually in tenths of a cc, or single grains, until the lather is permanent. The hardness is expressed in degrees. However, when the hard- ness is more than 16°, it is not possible to estimate it in this § 319.; 657 way with accuracy, and tne water under examination must be diluted with distilled water to double its bulk, and then the same quantity as above recommended taken for the estimation : in this case the number of degrees found must, of course, be multiplied by 2. " B. Hardness after Boiling.— K quantity of water, precisely the same in bulk as iii the former experiment, is boiled briskly for half- an-hour, and made up to the original bulk with distilled \yater; filtered, cooled, and treated with soap solution, exactly in the same way as in the previous case.* (10.) Alkalinity. — The alkalinity of water is best taken by using ns an indicator an alcoholic solution of tincture of cochineal, which is not afiected by carbonic acid, and sti-ikes a beautiful crimson purple colour with a trace of alkali, a reddish-yellow with acids. 100 cc. or more are placed in a tall cylinder of colourless glass, and a decinormal hydrochloric acid is run in, drop by drop, from a burette until the colour changes to a yellowish hue. The result is ex]>ressed in terms of carbonate of lime, each cc. of decinormal acid equalling 5 mgrms. of carbonate of lime. (11.) Or gallic Analysis of Water : Estimation of Organic Carbon and Nitrogen. — (1.) Carbon. — There are four main ways in which the carbon in a water residue is estimated — (1.) as gas; (2.)gravi- inetrically; (3.) nephelometrically ; (4.) indirect methods. (1.) Carbon as Gas. — The first method(which consists in burning up the carbon into carbon dioxide, and estimating both it and the nitrogen in a suitable gas-apparatus) we owe to Dr. Frankland, who proposed and practised it as eai'ly as 1867. FranMand's Combustion Process. — A quantity of water, varying from 100 cc. to a litre, accox-ding to the amount of impurity sus- pected from other determinations [especially of the free ammonia] is evaporated to dryness with special precautions. These precautions are mainly two — (1.) The protection of the sample from dust during the evaporating process, and (2.) the destruction of carbonates, nitrates, and nitrites, which, it is scarcely necessary to say, would greatly * A different method of determining hardness has been proposed by Mr. Hehner (Analyst, May, 1883). 20 cc. of normal sulphuric acid are made up to 1,000 cc, and a solution of sodic carbonate (l"Oii : 1000) is pre- pared, an equal volume of which exactly neutralises the acid. 1 cc. of the acid neutralises 10 mgrms. of calcic carbonate. 100 cc. of the water are tinted with cochineal, phenacetoline, or methylorange, and titrated in the usual way with the acid. Each cc. used indicates one degree of temporary hardness. To another 100 cc. a measured quantity of the sodic carbonate solution is added, more than enough to decompose the whole of the soluble lime and magnesian salts. The mixed solutions are evaporated in a platinum vessel to dryness. The residue is extracted with a little hot water, filtered and titrated hot with the standard acid. The alkali added, minus the acid used, indicates the permanent hardness as CaCOj. 658 foods: their composition and analysis. [§ 319. interfere with the results, and indeed render them vahieless. Small quan- tities of water, such as sewage and the like, can be evaporated under any- improvised cover, but for larger quantities I)r. Frankland recommends a self- filling circular water-bath, on the top of which rests a flanged copper capsule, serving as support to a thin glass dish, in which the evaporation of the water takes place. The dish is protected from dust by being covered by a tall glass shade, such as is used for statuettes. The bulk of the water for eva- poration is contained in a flask, to the neck of which is a ground glass tube bent appropriately. The flask, when filled with the water and connected with this tube, is by a quick movement inverted, so that the end of the tube rests on the glass dish. A little above the end of the tube there is a short side tube bent at right angles, of smaller diameter than the tube itself, the effect of which is that directly the water in the dish falls below the little angle of this tube, air bubbles up into the flask, and more water runs into the dish. In this way the evaporating dish is kept at a constant level until the whole of the water is used up. The steam condenses on the inside of the glass shade, and collects in the copper capsule underneath the glass dish, and is fiuall}' conducted away by a piece of tape which passes over the copper lip of the bath. The evapoi-ation of a litre of water takes about twenty-six hours, but with proper arrangements it is continuous, and when once started requires no supervision. The evaporating time is really a small matter, for the analyst can begin it one morning, and it will be ready the next. Before the water is submitted to evaporation it is boiled briskly with 20 cc. of sulphurous acid; or if previous estimations have shown that there is a lai'ger quantity of nitrogen as nitrates and nitrites than 5 per 100,000, a larger amount of sulphurous acid must be added. To ensure the destruction of nitrates, a drop of ferrous chloride is added to the first dishful of water. Lastly, in dealing with waters deficient in carbonates (in which case the sul])hurous acid, when oxidised to sulphuric, might not offer a sufficient base for combination, and therefoi-e there might be some destruction of the organic matter), 1 or 2 cc. of a saturated solution of hydric sodic sulphite are added, which will give any sulphuric acid, other- wise free, sufficient base for combination. When the evaporation is complete, the next step is to remove the residue from the dish and burn it up in a vacuum with oxide of copper. To avoid this removal, Dr. Dupve has proposed and used a collapsable silver dish : the water in this disli is evaporated down in the usual way, and then the dish can be rolled up and thrust into a combustion tube. If the analyst does not use the silver dish, the residue must be removed by the aid of a flexible spatula, and mixed with copper oxide. The combustion tube, 18 inches long and of rather narrow bore, is cleansed and dried, and charged by the aid of a small metallic scoop, first with a little coarse oxide, after which the residue is mixed with oxide ; lastly some more oxide is added, and in front of this is placed a roll of copper gauze which has first been oxidised in air and then reduced in hydrogen. The usual precautions in filling a combustion tube for organic analysis are, of course, to be strictlj^ observed. The tube is now placed in a combustion furnace, exhausted of all air by a Spi-engel pump, the tube made gradually red-hot, and the gas finally trans- §319.] ferred to a gas apparatus. WATEP. 659 The present writer has forsaken the use of Sprengel pumps, and for this and all other vacuum operations employs the mercury pump de- scribed at page 70, the figure of which is here repeated. The combustion tube is attached by a short piece of pressure tubing direct to the bent tube at Z, made perfectly vacuous, and when no air is delivered into B the stopper, F, is replaced by the tube SST. After having filled the thread of SS'P with mercury, and again seen that there is no air in the apparatus, it should be left for a short time to ascer- tain whether the single joint is. sound. Any leakage is dis- covered by the mercury column sinking in AA. All being sound,, the point P is placed under any eudiometer or gas-apparatus which the analyst may have, and the combustion proceeded with^ the gas finally being pumped into the measuring apparatus by opening the stopcock S', and working the reservoir R up and ^^g- 73- down. The author also uses a gas apparatus somewhat diflferent from that generally employed. This appai'atus consists (see fig. 74) of a reservoir, A, attached by an india-rubber pressure tube to a glass tube, B, which in its turn is connected with the steel block invented by Mr. J. W. Thomas.* The barometer tube, E, pos- sesses no stopcock, it being unnecessary; but it is filled perfectly with mercury, and then stoppered with caoutchouc and a bit of glass rod, and finally sealed from any air-leakage by a little cell filled with mercury (S). The laboratory tube, V, is a glass cylinder of ^ diameter and 5 in. in length, and provided with several divisions at equal distances. It is secured by a suitable clamp in the mercury trough [this clamp is omitted in the dia- gram], and when it is wished to connect it with the reservoir and barometer this is easily effected by slipping it on to the end of the * Joxirnal of the Chem. Society, May, 1879, p. 218. 660 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 319. tube, I, which is fitted securely and permanently into the bottom of the mercury trough, and has been ground so as to fit the end of the laboratory tube with a perfect yet easy joint. V is also jacketed, as indicated by dotted lines. The jacket is a short cylinder, closed at the top by a caoutchouc stopper, through which pass two tubes, the one connected with india- rubber tubing from the barometer jacket, the other connected with a waste pipe. There is no 2)ractical difi'erence as to the particular course for the water to take. It laay enter the top or the bottom of the jackets, but there must be a con- stantly running stream. It is necessary, first of all, to find the exact milli- metre division of the barometer tube, wliich each division of V corre- sponds to, and this is efi"ected by directing a low power telescope, or, as for that, a tube to eacli division provided with Fi.g. 74. It is also necessary to spirit level, know the exact capacity of V, which is obtained by filling it with water, and allowing air to gradually displace the water, weighing the different fractions of the displaced water at each division. To make an ordinary analysis of a gas containing nitrogen and carbon dioxide, the gas is collected or transferred into V ; V worked on to I ; the reservoir lowered, after opening the way in the steel block, to both barometer tube and laboratory vessel, until any convenient division in V is reached (a good stream of water running all the time through the jackets) ; the height of the bai'ometer tvibe is now read, the tempei'ature of the water noted, and the usual calculations made. To absorb the carbon dioxide, V is carefully removed from oft" I, of coui'se taking care in the removal that the open end of V is not lifted 661 above the mercury. A .small bit of potash is now melted on to a platinum wire, moistened and thrust up into the gas ; it remains there until by shaking no further absorption is observed. The wire is then withdrawn, and V worked on to I as before, and the gas measured. If the laboratory vessel is furnished with platinum wires near the top, the apparatus can also be used for analyses by explosion. The writer prefers to absorb carbon dioxide in the old-fashioned way, by a piece of potash, so as not to soil the mercury too much by liquids ; but when it becomes a ques- tion of the use of alkaline pyrogallate or Kordhausen sul})huric acid, it is better not to use the laboratory vessel described, but to slip on to the end of I a small piece of rubber tubing, so as to make the large end of an ordinary burette fit mercury- and air- tight. The upper and smaller end of the burette is provided with good pressure tubing provided with a clamp, and by means of a bit of thick- walled capillary glass tubing connected with the pipette figured. The pipette consists of two bulbs, A and B. C is a thick- walled glass tube, with a capillary bore ; at X is the India-rubber con- nection. Befoi'e commencing an ab- sorption, A must be filled through B with the reagent ; it is then, when properly connected, easy to drive the gas over into A, and also at the last a thread of mercury sealing C. The jiipette may now be disconnected and well shaken without any loss of gas, and absorption is far more rapid than when liquid reagents are applied in the ordinary manner. An analysis of a sample of air, and one of a gas con- sisting of carbon dioxide, nitric per- oxide and nitrogen, may be cited as an example. Air. — A sample of air brought to the tenth division of a burette fitted on to I, barometer tube reading 900 mm., but the barometer reading corresponding to division 10, is 325 mm. Therefore this has to be subtracted : 900 - 325 = 575 mm., which is the pressure of the total gas. On taking the burette ofi* I, and inserting a moist stick of potash, and again reading at the same division, the barometer reading is now 899-8 mm. ; this subtracted from 900 gives, as the tension of the carbon dioxide, •2 mm. Lastly, on impelling the gas into the pipette (fig. 75), and submitting it to the action of alkaline pyrogallate for two Fig. 75. 662 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 319. hours with frequent shaking, and measuring at the same divi- sion, the barometer reading is 780-9, which, subtracted from the former reading gives as the pressure of the oxygen 119 mm. Hence Pressure of carbou dioxide, .... "2 ,, oxygen, ...... 119"0 ,, nitrogen, 455*8 100 = = carbon dioxide 20-0 = •03 100 = = oxygen 11900 ' 575 = = 20-69. Total pressure, . . . 575*0 Since the temperature was constant throughout, if the volume in percentages only is required, the calculation is as follows : — (1.) 575 : -2 (2.) 575 : 119 That is, the air contained -03 per cent, of carbon dioxide, and 20*69 per cent, of oxygen. In a determination of the carbon and nitrogen of a water residue (from a litre), the following is an example of the method and of the numbers obtained : — The gas was pumped out by the mercury-pump direct into V, V was then fitted on to I, and measured at the fourtli division ; the pressure of the mercury as read on the barometer tube was 850 mm.; but since the division itself equalled or corresponded to 284, the total pressure of the gas was 850- 284 = 566 mm. On now absorbing by a stick of potash, the pressure was found to be 534, therefore the ten- sion of the carbon dioxide was 850-584 = 266. Two bubbles of pure oxygen were now added, and the gas, which immediately became of a red colour, su.bmitted to the action of alkaline pyi'ogallate ; after this operation the barometer reading was 574, and therefore the pressure of the nitric peroxide was 584-574 = 10 mm. We have, therefore, the following deter- minations : — Division 4 = 23 5 cc. Temp. = 11° *5 mm. Tension of nitrogen and nitric oxide, . . . 300 „ nitrogen, 290 „ the three mixed gases 566 These operations have furnished three uncorrected volumes of gases — § 319.] WATER. 663 A. Volume of the three mixed gases. B. Volume of nitric oxide and nitrogen. C. Volume of nitrogen. The volumes must, therefore, all be reduced by the usual cal- culations to 0° temp, and 7G0 mm. pressure. From the corrected volumes the quantities of carbonic dioxide and nitrogen may be reduced as follows : — A - B = volume of CO2 B-C ^ B + C , , . — Q — + U — — 2~ = volume of mtrogen. From these corrected volumes of nitrogen and carbonic dioxide, the weights of carbon and nitrogen can be obtained by calcula- tion or by tables. There is, however, a far simpler means of ariiving at the desired result by the aid of the following data : — 1. The weights of carbon and nitrogen contained in equal volumes of carbon dioxide and nitrogen gases, measured at the standard temperature and pressure, are to each other as 3 : 7. 2. The weights of nitrogen contained in equal volumes of nitrogen and nitric oxide are as 2 : 1. Hence, if we assume that for the purpose of calculation the gaseous mixture consists entirely of nitrogen, and that two successive portions of the nitrogen are removed from it by the reagents ; then, if A be the weight of the total gas calculated as nitrogen, B the weight after absorption of the first portion (COo), and C the weight after the absorption of the second por- tion (NoOg) ; further, if x and y represent respectively the weights of carbon and nitrogen contained in the gaseous mix- ture, then the following simple equations express the values of X and y : — _3(A-B) C + B In the example given the calculation is as follows : — Log. of div. 4 (23-5 cc.) = 1-37106 Log. of 566, = 2-752S1 Log. from table LV I . , p. 697, c orresponding to 1 1 ° -5, = - 6 -20029 _ -2-32416" •02109 Log. div. 4, = I-37IO6 Log. of 300, = 2-47712 Log. from table LVL, p. 697, corresponding to ll°-5, = - 6-20029_ -2-04847 = "°"^^ 664 FOODS : their composjtion and analysis. [§ 319. C. Log. div. 4, , . . , ^ 1-37106 Log. of 200, . . . . = 2-46239 Log. from Table LVl., p. 697, 1 _ - 6-20029 _ corresponding to ll°-5, . J ~ - 203374 ~ "*" n u 3( -02109 - -01187) ^^,^^ Carbon, = -^^ -— -00438 ,,., -01187 + -01080 ^,,__ Nitrogen = = -01133 Or the water contains in 100,000 parts -438 carbon, 1-133 nitrogen. Blair's Metlwd of Moist Combustion. — 250 cc. of water are acidified with 2 cc. of pure, strong sulphuric acid when the water has been concentrated to 100 cc. ; all COj from carbonates is then boUed off by concentrating to 50 cc. A globular receiver with two necks, one of which can be adapted airtight by a rubber connection to the retort, is now charged with 10 cc. of permanganate and 100 cc. of water. The receiver is connected with the retort, and to the second tubular portion is adapted an indiarubber tube with clip. The liquid in the retort and that in the receiver are each made to boil by means of two Bunsen burners ; after the steam has issued from the exit tube for a few minutes, the exit tube is clipped, and at the same moment the flames removed. The condensation of the steam produces a vacuum, as shown by the flattening of the rubber tube. By tilting the apparatus a little, about 30 cc. of the permanganate solution is allowed to flow into the retort from the receiver, and the retort heated gently ; distillation from the retort into the cooler receiver at once commences, and ultimately the acid, becoming concentrated, oxidises all the organic matter; the resulting solution should be colourless, but, if it is not so, a little more permanganate should be transferred as before, and the operation repeated until a colourless liquid results. The gases now in tlie apparatus are carbonic acid, oxygen, and chlorine, and it is obvious that, by means of a Sprengel or mercury pump, they could be transferred to a suitable gas measuring apparatus, and the quantity of COg ascertained after absorption of the chlorine by direct measurement ; Blair, however, prefers to estimate the CO2 by titration. 10 cc. of a 10 per cent, solution of ferrous sulphate are passed into the receiver by adjusting the charged pipette to the exit rubber tube, and carefully loosening the clip, so as to allow no air to enter. The contents of the receiver decolorised by the ferrous sulphate are allowed to flow into the retort, the liquid heated § 319.] WATER. 665 nearly to boiling and then poured backwards and forwards from retort to receiver. In a few minutes all the chlorine is absorbed. A 6-oz. flask is prepared, having an accurately fitting rubber cork, through which passes a glass tube fitted with a rubber con- nection ; a little water is placed in this flask, the water boiled to expel air, and the rubber tube clipped while the water is boiling ; the flame is removed, and the flask cooled. Into the vacuous flask 10 cc. of barium hydrate solution, 0'315 per cent., tinted with phenolphthalein are run in through the indiarubber tube by placing the end of the charged pipette in the end of the rubber tube and carefully loosening the clip ; the glass tube is then pushed down until the lower end nearly reaches the bottom of the flask. The flask is now connected with the receiver by the indiarubber connections, the clips loosened, and the liquids in receiver and retort boiled, the flask being cooled in water ; if the rubber tube remains flat, this is a sign that no air has leaked in, and that the steam has not sufiicient tension to do any damage. Boiling from five to seven minutes expels all COo ; the flames are then removed, the rubber tubes clipped, the flask finally disconnected, and the contents titrated with oxalic acid. The exact amount of c. n. oxalic acid required to neutralise 10 cc. of the baryta water deducted from that required to neutralise the same quantity in the flask, gives the requisite data for calculating the CO.,. 10 cc. of c. n. oxalic acid = 1-2 mgrm. of carbon. The AufJior's Metliod of Moist Combustion. — Half a litre of the water to be examined is first saturated with SO^, and transferred through the thistle-headed funnel (fig. 76) of A into the flask, A. This is a strong Florence flask of about 600 cc. capacity, having a side tube, the side tube being connected with a water pump and a mercury apparatus as follows : — A smaller flask, capacity about 70 cc, with a T-side tube, is connected with the side tube of A by means of a rubber cork. There is a Bunsen valve at v, so that no gas or liquid can run back into A ; one limb of the T-piece is connected with the water pump, and the other with a gas burette graduated in cubic centimetres and tenths of cubic centimetres charged with mercury, and provided with a pressure tube and a stopcock. There are also suitable clamps on the rubber parts. The measuring burette is first filled with mercury by raising the pressure tube, P, and, when tilled, shutting off by means of the stopcock at s ; the clamps, t and g, remain open. The flask, A, is placed on a sand-bath, the water pump set going, and the water boiled so as to get rid of free SOo ; this only takes a few minutes. Next, about 20 cc. of clear saturated baryta water 666 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 319. are transferred by means of the thistle funnel into A, without letting any air in, and this is followed by 10 cc. of 1 per cent, pei'manganate. The alkaline solution is boiled down in a vacuum almost to dryness. The flask is then raised, cooled slightly, and 5 cc. of organically pure sulphuric acid diluted with about 20 cc. of water are transferred through the funnel in the same way. The clip, t, on the tube leading to the water pump is screwed down, and the apparatus put in connection with the gas burette by opening, s. The flask, A, is now lowered on to the sand-bath and allowed to boil ; nearly all the water condenses in B, which, at the beginning of this part of the operation, is almost empty, the condensed vapour from the former operations having passed away to a receiving flask connected with the water pump. The liquid is distilled to dryness and allowed to fume for about an hour. With ordinary waters the acid fumes are arrested in B. At the end of an hour the flask is removed from the sand-bath, allowed to cool slightly, and then recently boiled solution of ferrous sulphate run in through the thistle funnel. This is boiled for a few minutes, § 319.] WATER. 667 and, lastly, heat is applied for a few minutes to B. All tlie carbon dioxide is now in the gas burette, s is closed, the flame removed, and the gas burette detached from the apparatus ; it is allowed to cool, and the gas carefully measured by bringing the meniscus of the pressure tube, P, on an exact level with that in the measuring burette and reading the number of cubic centimetres. The height of the barometer and the temperature is also ascertained. The gas is now transferred to the bulb pipette charged with potash, figured at p. G61, the bulb pipette is shaken, the unabsorbed gas drawn again into the measuring burette, and the gas again measured. The contraction will be equal to the amount of COo. The whole readings must be reduced to normal temperature and pressure, and the volume of CO., translated into parts by weight of carbon. If, instead of measuring the carbon dioxide as a gas, the analyst prefers to absorb it in baryta water, the burette is replaced by an absorption apparatus consisting, first, of a bulb containing potassic permanganate solution, and then of absorp- tion vessels charged with clear baryta, the last vessel being connected with the water pump ; the carbonate of baryta can be filtered off under xylene, washed with hot water, dissolved in hydrochloric acid, evaporated to dryness, and converted into barium sulphate by adding a little sulphuric acid to the dry residue and driving the excess of acid off by heat. The advantage of this method of making a moist combustion is obvious. The permanganate, by acting first in alkaline solution in a vacuum, at once fixes those volatile organic matters which in boiling down would otherwise be lost ; while (except for a few minutes) the water is constantly under the influence of an oxidising medium, which is not the case with other processes. Gravimetric Estimation of Minute Quantities of Carbon. — Drs. Dupre and Hake in 1879* published a method of gravi- metrically estimating minute quantities of carbon by burning up in a current of pure oxygen, and then absorbing the CO., thus produced in baryta water, and converting the baryta carbonate into baric sulphate. A product was thus obtained which weighs 19-4 times as much as the carbon originally present. The details of the process are as follows: — A combustion tube, open at both ends, and about 24 inches long, is drawn out and bent downwards at one end at an angle of 120°, so that it may be conveniently attached to a Pettenkofer's absorption tube, the other end being connected, by means of a caoutchouc stopper and * Journal of Chemical Society, March, 1879. 668 FOODS : their composition and analysis. [§ 319. glass tubing, with an oxygen reservoir. This combustion tube is filled half way from the bent end with granulated cupric oxide, which may conveniently be held in position either by plugs of asbestos or by platinum wire gauze, or by a combination of both. The connection with the oxygen reservoir being then made, the greater part of the tube is heated to redness, with the ordinary precautions, and a stream of oxygen (which is first con- ducted through a long tube containing caustic potash) is jiassed over the glowing oxide of copper until the issuing gas ceases, after long bubbling, to cause any turbidity in the bright baryta water. As soon as this point is reached, the portion of the com- bustion tube preceding the layer of cupric oxide is allowed to cool somewhat, and the tube is now ready to be connected with the absorption apparatus. The clean absorption tube is care- fully rinsed with water, and is clamped in front of the furnace in such a manner that its bulb end is somewhat higher than the end to be connected to the combustion tube. Both ends must be provided with convenient stoppers, consisting of short pieces of caoutchouc tubing closed with a small piece of glass rod. The stoppers being removed, air, which is first caused to pass through a tube containing caustic potash, is pumped through the tube for about two minutes, and it is then filled with baryta water as follows : — The baryta water (of strength 1-5 per cent.) is kept in a sufficiently large stock bottle, provided with a caoutchouc stopper, thi-ough which pass two bent glass tubes, the long one for syphoning, the shorter, to which a potash tube is attached, being connected with a small hand-bellows. In filling the absorption apparatus, the longer syphon tube is connected with it by means of flexible tubing, and the baryta water is forced over by gentle pressure of the bellows, the bulb end of the absorption apparatus being provided with a potash tube. As soon as the absorption apparatus is half filled, the flow of baryta water is arrested ; the ends of the Pettenkofer tube are immediately closed by its stoppers, and it is now ready for use. By these means the tube is filled with perfectly clear and bright baryta water. The absorption apparatus is now connected with the combustion tube, and the combustion proceeded with. The silver dish containing the water residue having been inserted just behind the copper oxide, it is burnt in a slow current of oxygen, and the carbon dioxide is absorbed and converted into baric carbonate in the absorption tube. In order to filter off and convert the baric cai'bonate, a funnel and filter are arranged to stand over a beaker containing a layer of caustic potash solution at the bottom, the whole being covered by a bell jar, which itself stands in a layer of caustic potash solution. The § 319.] WATER. G69 mouth of the bell jar, which is immediately over the funnel, is closed by a thick caoutchouc cap with two narrow openings, one of which is provided with a caustic potash tube. (Soda lime apparently answers equally well.) Tlie other, wliich is tem- porarily stoppered, contains a straight glass tube, placed imme- diately over the filter so that, after the whole arrangement has been left some time to itself, in order that all enclosed air may be free from CO.,, direct connection may be made with the Pettenkofer tube by means of flexible tubing sufficiently long to admit of some slight freedom of action. Filtration may thus be carried on without danger of CO, being introduced from the atmosphere, the additional precaution being taken of compelling all air which passes through the Pettenkofer tube during this process of filtration, to pass through a tube containing caustic potash attached to the tube itself. The washing of the preci- pitate in the tube and on the filter is effected almost entirely with boiling water, which lias been previously saturated with carbonate of barium [solubility 1 in 15,000], but finally with a small quantity of boiling distilled water. After complete wash- ing, the tube is disconnected, and the filter ultimately rinsed round, while still under the bell jar, by means of the long tube already mentioned, and which, when not clamped, may be moved freely in all directions. The bell jar is then removed, and the pre- cipitate is rapidly washed together into the bottom of the filter. The Pettenkofer tube, which may contain minute particles of baric carbonate not removed by the washing, is rinsed twice with small quantities of dilute pure hydrochloric acid (about 1 in 50), and finally with distilled water : the rinsings are poured on to the filter on which the greater mass of baric carbonate is already collected. The filter is further washed with dilute hydrochloric acid, and finally with distilled water : and the whole of the solution of baric chloride so formed is cai-efuUy collected in a small beaker. The quantity .of such solution need not exceed 50 cc. This solution of chloride of barium has next to be evaporated, which is best done in a platinum vessel on the water-bath. It is then transferred, when greatly decreased in bulk, to a much smaller platinum dish, weighing about 5 grms., and finally evaporated to dryness after the addition of a few drops of pure sulphuric acid. The dish and its contents have then to be ignited, the residue moistened with a drop of nitric acid and re- dried, and the whole re-ignited and weighed to conclude the operation. The amount of carbon present is obtained by dividing the weight of the baric sulphate by 19-4. Nephalometric Method. — This ingenious metliod we also owe to Dupre and Hake. The carbonic acid resulting from the combustion of an organic residue is passed into perfectly pure 670 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 319a. clear solution of basic lead acetate, and the turbidity produced is imitated by known weights of CO., ; in fact, the operation is a colour method conducted on the same principle as " Nesslerising," with this important difference, that no success will be obtained unless there are special precautions taken to prevent the con- tamination of the solutions by the breath and air, itc. § 319a. The Estimation of Organic Nitrogen after KjeldahVs Method. — Haifa litre of the water is placed in a retort and the free ammonia distilled off. Then 5 to 10 cc. of diluted sulphuric acid are added, and the water concentrated down on a sand- bath until the acid fumes. The acid is allowed to fume for about half an hour, or until it is almost colourless. The acid solution is then cooled, diluted, alkalised with pure soda lye, and the liquid distilled, the alkaline distillate being neutralised with decinormal sulphuric acid, each cc. of which is equal to 1*4 mgrm. of nitrogen. This simple process is applicable to most pure waters containing but little organic matter and feeble nitrates. On the other hand, it will not give accurate results with waters containing much nitrate or much organic matter. In such a case the following is the best method :*— Half a litre, as before, is taken. The water (after getting rid of the free ammonia) is saturated with SO2, and a drop of iron chloride solution added ; it is then gently heated for about twenty minutes ; and is next boiled down to about 20 cc. To this residue is added 20 cc. of sulphuric acid containing 4 grms. of phosphorus pentoxide and then 0*12 grm. of anhydrous copper sulphate and 5 drops of platin chloride solution. The contents of the flask, closed by a glass marble, are heated gradually to a gentle boil, and the heating continued until the fluid remains of a green colour. After cooling the acid fluid is diluted, alkalised by ammonia-free soda solution, a little granulated zinc added, and the whole distilled into a measured volume of deci- normal sulphuric acid. The decinormal acid arrests any ammonia, and on titrating the distillate with d. n. soda, there will be a loss of acidity proportionate to the ammonia, fi'om which (as before) the amount of nitrogen can be calculated. There has been found a slight practical difficulty in thus titrating ammonia with the greatest accuracy, and Kjeldahlf has, therefore, recommended the utilisation of the reaction which takes place between a free acid and iodide and iodate. This reaction (denoting free acid by HR) takes place as follows : — 6HR + SKI + KIO = 6KR + 3HoO + 61. To the acid distillate is added 0-4 grm. of potassium iodide * Ulsch, Zeitschrift f. analyt. Chemie., xxv. 579. t Kjeldahl, ib., xxii. 366. § 319a.] WATER. 671 aud 0-1 grm. of potassic iodate ; after standing two hours the iodine set free is estimated by means of a decinormal thiosulphate solution, the strength of which has been checked and adjusted by the aid of a decinormal iodine solution. An example will make the calculation clear. The distillate from half a litre of water was submitted to Kjeldahl's process, and was received in 30 cc. decinormal sul- phuric acid. Potassic iodide and iodate were added thereto (as above described), and the mixture allowed to stand two hours ; at the end of that time the free iodine was estimated by means of a decinormal thiosulphate solution, using starch as an indi- cator ; i'5-9 cc. of thiosulphate were used. Since 30 cc. of thiosulphate are equivalent to 30 cc. of decinormal acid, it is clear that 30 - 25-9 — that is, 4-1 cc. of free acid— have been saturated with ammonia ; hence the distillate contains 4-1 cc. X r4 mgrm. = 5-74 mgrms. nitrogen, and the water contains 11-48 mgrms. per litre or 11-48 parts per million of organic nitrogen. (12.) Mineral Analysis of Water * — Ordinary drinking water holds dissolved but few saline matters, and when an analyst has determined chlorine, nitrates, sulphates, phosphates, and carbonates, and also lime and magnesia and alkalies, he will usually find, on adding the several amounts together, that he gets numbers very nearly equal to the solid saline residue. An excellent method of approximately estimating the various saline con- stituents of a water is to evaporate clown to dryness a known quantity, then to treat the residue with a little hot water, which will dissolve all the * A method of determining calcium and magnesium has been described by Professor C. L. Bloxam [Chem. News, 1886) ; it depends upon the precipitation of calcium and magnesium as ammonio-arsenates, and is specially applicable to their separation from strontium salts. The deter- mination of calcium as ammonio-arsenate in ordinary drinking waters lias some advantage — the precipitate is as nearly insoluble in water as calcium oxalate, and being highly crystalline is not liable to run through the tilter — the formula of the precipitate dried at 100° is Ca5NH4H5(As04)c.3H„0, and 100 parts are equal to 20 of calcium or 50 calcio-carbouate. If a rapid determination be desired, arsenic acid is added to ^ of a litre in water, which is then strongly alkalised by ammonia. The mixture is well stirred, and allowed to stand for ten minutes, the precipitate is then collected on a weighed filter, washed with anunonia water (S"5 per cent.), and dried at 100°. The gain in weight represents the united magnesic and calcic ammonio- arsenates. The precipitates are now dissolved off the filter by acetic acid, and the calcium precipitated as oxalate by ammonium oxalate the solution boiled and filtered. On adding to the filtrate ammonium, the ammonio- magnesium arsenate is in this way reprecipitated, and may be collected on a weighed filter, washed with ammonia water, and dried at 100° ; its composition is (MgNH4As04iH^O) ; 100 parts are equal to 44-2 of magne- sium carbonate. 672 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 320. soluble salts out, but leave insoluble carbonates of lime and magnesia, and silica. In the soluble portion, the soluble salts of the alkaline earths and the alkalies are determined ; the chlorides, sulphates, and nitrates are estimated on the unconcentrated water by the processes already detailed. It is also always open to make the aualysis in the old-fashioned way — that is, to evaporate down a large quantity of water, to separate the silica by treatment of the ash or residue with hydrochloric acid, and after separa- tion of the silica to divide the solution into three or four quantities, ia which sulphuric acid, lime, magnesia, &c., are determined by the ordinary methods. IV. Biological Methods. § 320, A. Microscopical Appearances. — To make a microscopical examination of water, it is necessary to collect the sediment or deposit which falls to the bottom of the vessel in which the water stands. A convenient way of doing this is to use the author's tube (fig. 77), wliich holds a little more than a litre. The little glass cell, C, is adjusted to the pipette-like end, the rod is removed, and after introduction of the water the tube is covered and set aside for twenty-four hours. At the end of that time any deposit will have collected in the glass cap. On now carefully inserting the rod-like stopper, the cap or cell can be removed with great ease, and its contents submitted to microscopical examination. With very pure waters merely a little sand or formless debris collects in the cap, and there is no life. If, however, in the first place eight or ten gallons are allowed to deposit in a capacious vessel, most of the water run ofi", and then the last litre rinsed into the tube, in nearly every case there may be a few life-forms and sufiicient matter collected to give definite results. It need scarcely be said that an opinion must not be formed upon a micro- scopical examination without taking into account the amount of water from which the sediment has been collected, and a definite quantity should be generally Fig. 77.- agreed on by analysts. The Sedgivich-Kafter Method* — This ingenious method of obtaining a quantitative estimation of organisms and objects in water is in use by the Massachusetts biologists. A brass gauze stop is put in the mouth of a funnel, and on this stop is picked a layer of shar)) quartz sand; the sand grain should pass through a sieve 80 meshes to the inch, but not through a sieve 100 to the inch. On to this sand is poured a con- * Massachusetts State Board of Health lieport. Boston, 1890. § 320.] WATER. * 673 venient quantity of the water (say 500 cc), well shaken, so as to get a fair sample. The sand and organisms, after filtra- tion, are washed into a test-tube, with from 3 to 5 cc. of distilled water. The tube is shaken thoroughly, the sand allowed to settle for a moment, and the water quickly decanted. This procedure leaves the sand, but transfers most of the organisms. 1 cc. of the liquid is placed on a counting plate, consisting of a shallow cell with a brass border, the cell holding exactly 1 cc. A fractional part of the field is now observed, and the organisms counted therein ; in order to do this, the slide is ruled into millimetre squares, or a metal disc is fitted into the eye-piece with a square hole cut in its centre, the area of the square hole being such that with the powers used it just covers a square millimetre. The Massachusetts biologists have by this method examined an enormous number of waters ; some having been examined on several days in each month in the year. The method of record- ing and tabulating the results are shown in Table LIII6., giving the biological results of the water supply of Nantucket. This water supply, owing to the increase of Anaboena (one of the blue-green algpe), in the months of August, September, and October smells and tastes offensively. 3/r. Bibditi's Process. — A better method than the Sedgwick- Kafter process is that which has been invented by the chemist to the London County Council (Mr. Dibdin). A litre of the water (or less, should the water contain much suspended matter) is filtered through hard filter paper. The deposit is washed off the filter paper into what is called a " micro-filter." The micro- filter is prepared as follows : — A piece of clean combustion tubing is drawn out into a capillary tube of a diameter of 2 mm. The open small end is plugged by a paste made of equal parts of air-dried clay and kieselguhr. This plugged end is dried in the Bunsen flame, and ultimately heated to redness. The residue from the water is now placed in this micro-filter, and the micro-filter is fitted into any convenient flask or bottle by means of an india-rubber plug attached by a side tube to a good water-pump, and the superfluous liquid drawn through, until only about 1 to 1'5 cc. remain. The sediment and suspended matter are thus collected in the form of a compact cylinder just above the porous filtering substance, and are now carefully measured. The results being expressed in millimetres per litre. The cylinder of deposit is removed by scratching the tube with a sharp file about half an inch from the filter plug and breaking it off. A platinum wire is pushed in so as to loosen the deposit from the porous plug, 44 674 FOODS : THEIR COMPOSITION AND ANALYSIS. [§ 320. 5S ^ 2 i 2 S « t-t < ^ r b £ ai i O o o o s o (M OJ o o as ^ U3 »o o o ^ o - ^ rH - 05 o 05 o §3o i o t-" s4 o ^ c. 05 ■"^ CO (M CO o o o o ■* 3 o o r^ CO (M o O !N o o -rt" ^ o c4~ CO to o o ^ (M to • • • • • • • • t^ CO ^ 0) ^- «? \}i g ^ oT „ Z 8J 3f 1 i c8 1 i 1 i i 3 s "■ i ft i 8 <5 1 la oio i |oog,oo^o ^^^^^^ g^O (N oo^ 1 J^OOOOOOOO C0|f500-H0 ooo o OO cooooooocoo c-ooo-io ooo -o» g Ooooooooc-io oooooo ooo 1 o g2 s?l 1 CM 000000(M0 rji (N (M O O O O OO gs§ ^ ! 00 OCO O OOOOOOOO O OOOOO O OO 1 o o 1 ^ ^ i O OOOOOOOO O OOOOO o OO ■ o OO g O OOOOOOOO O OOOOO 2 '^2 • 2 2- * i ca oooooocoo ^ eoo-HOo o "^o O OO i O OOOOOOOO ^ ^oo-co g °§ %'s § ^ 05 OOOOOOOO -* rwO(NO-H O OO gsg 2 OOOOOtNOTt'OlO 00^O,000. Solution 100,000. Solution 100,000. Solution 100,000. •7 •00 4 6 5^43 8-5 11^05 123 16^90 •8 •16 •7 •57 •6 •20 •4 17 06 •9 •32 •8 •71 •7 •35 •5 •22 ro •48 •9 •86 •8 •50 •6 •38 •1 •63 5^0 6-00 •9 •65 •7 •54 •2 •79 •1 •14 90 •80 •8 •70 •3 •95 •2 •29 •1 •95 •9 •86 •4 Ml •3 •43 ■2 1211 13^0 18^02 •5 •27 •4 •57 •3 •26 •1 •17 •6 •43 •5 •71 •4 •41 •2 •33 •7 •56 •6 •86 •5 •56 •3 •49 •8 •69 •7 7^00 •6 •71 •4 •65 •9 •82 •8 •14 •7 •86 •5 •81 2-0 •95 •9 •29 •8 13 •Ol •6 •97 •1 2^08 6 •43 •9 •16 •7 19^13 •2 •21 •1 •57 10 •31 •8 •29 •3 •34 •2 •71 •1 •46 •9 •44 •4 •47 •3 •86 •2 •61 14^0 •60 •5 •60 •4 8^00 •3 •76 •1 •76 •6 •73 •5 •14 •4 •91 •2 •92 •7 •86 •6 •29 •5 14^06 •3 20 08 •8 •99 •7 •43 •6 •21 •4 •24 •9 312 •8 •57 •7 •37 •5 •40 3-0 •25 •9 •71 •8 •52 •6 •56 •1 •38 7^0 •86 •9 •68 •7 •71 •2 •51 •1 9 00 110 •84 •8 •87 •3 •64 •2 •14 •1 15 00 •9 2103 •4 •77 •3 •29 •2 •16 15 •19 •5 •90 •4 •43 •3 •32 •1 •35 •6 4 03 •5 •57 •4 •48 •2 •51 •7 •16 •6 •71 •5 •63 •3 •68 •8 •29 •7 •86 •6 •79 •4 •85 •9 •43 •8 10^00 •7 •95 •5 22^02 4 •57 •9 •15 •8 1611 •6 •18 •1 •71 8^0 •30 •9 •27 •7 •35 •86 •1 •45 12^0 •43 •8 •52 •3 5^00 •2 •60 •1 •59 •9 •69 •4 •14 •3 •75 •2 •75 160 •86 •5 •29 •4 •90 § 323.] TABLE LVI. — Keddction of Cubic Centimetres of Nitrogen TO Grammes. 00 12562 log. ^j^ Q.QQ^g, , 760 for each tenth of a degree from 0° to 30° C. cc. T 0-1 0-2 0-3 0-4 0-5 0-6 0-7 0-8 0-9 6-21824 808 793 777 761 745 729 713 697 681 1 665 649 633 617 601 586 570 554 538 522 2 507 491 475 459 443 427 412 396 380 364 3 349 333 318 302 286 270 255 239 223 208 4 192 177 161 145 130 114 098 083 067 051 5 035 020 004 *989 *973 *. lithisis, 296. of cows suffering from pneu- monia, 295. of cows suffering from typhus, 728 Milk of goat, 257, 264. ,, of hippopotamus, 259. ,, of infants, 262. ,, of llama, 259. ,, of mare, 258. ,, of sheep, 259. ,, of sow, 260. ,, of woman, 232, 255, 256. ,, — Penalty for refusing to sell samples to inspector, 55, 70S. ,, —Preservation of, .317-320. ,, — Pro])agation of disease through, 298. ,, —Quantity yielded by cows, 326. ,, — Samples of, taken in transit, 55, 64, 70S. „ —Skim, .331. „ — Skimming of, 331 ; legal case, 3.34. ,, —Specific gravity of, 268. ,, standards, 309. 245, 246. — Estimation of, 139, 140, 284-280. ,, —Total solids of, 238, 268. „ tree, 262. Milking cows— Food of, 320. Millet, 177, 217. starch, 177. Millon and Commaille's lacto-pro- teine, 247. Millon and Kekule's analysis of bran, 181. Mitchell, John, 41. Mitscherlich's polarising sacchari- meter, 145, 146. Mogdad coffee, 441. Moissan's method of detecting borax, 312. Molasses, 158. Molybdic acid solution, 692. Morine, 101. Moritannic acid, 101. Mucedin, 185. Mucor mucedo, 202. Mulberry, 159. Muller'« pxperiments on cheese, 382. ,, process of milk analysis, 278. Muscarine, 111, Must, 545. Mustard, 593-602. ,, — Adulteration of, 599. „ —Chemistry of, 596. Mustard — Microscopical characters of seed, 593. volatile oil, 599. Muter's method of analysis of insol- uble fatty acids, 369, 370. Mutton fat— Melting-point of, 354. Myronate of potash, 698. Myrosin, 114, 597. Myrtle or Russian green, 113. Naphthalene red, 96, 112. yellow, 112, 116, 600. Naphthol yellow, 114. Naphthylamine test for nitrates, 632. ,, ,) for pepper, 610. Natal arrow-root, 172. Neatsfoot oil — Refraction of, 359. Negri and Fabris, examination of olive oil, 627. Nephalometric method for minute quantities of carbon, 669. Nessler & Earth, general process for analysing wine, 552. ,, solution, 693. ,, test in water analysis, 655. Nestle's biscuit food, 209. Swiss milk, 209. Neufchatel cheese, 380. Neumann. Caspar — Observations on milk, 236. Newton's patent process for preserv- ing milk, 318. New York Adulteration Act, 711. Night blue, 111. ISigrosine, 112. Nile blue. 111. _ Nitrates as evidence of watering wine, 561. ,, — Estimation of, in pepper, 608. ,, — Estimation of, in water, 6.32, 642-646. „ in milk, 292. Nitrites in water, 636-641, 646. —Tests for, 6.31. Nitro-benzine— Detection of, in bitter almonds, 619. Nitro colours, 109. Nitrogen, albuminoid — Estimation of, 124. ,, — Table for reduction of cc.'s to grammes, 697. ,, in bread, 199. ,, in cereals, 182. ,, in cocoa, 463. 729 Nitrogen in tea, 414. ,, in milk — Estimation of, 2SS. ,, in peas, 222. , in water— Determination of, 6G2-664. ,, — Non-albnminoid, 124, Nitroso-colours, 110. Normandy, Alphonse, 25, 41. Norwegian condensed milk, 209. Notices on shops, &c., relative to ptirity of articles, 51. Nuclein, 244. Nutmeg starch, 173. Oatmeal, 167, 175, 177, 210, 211. ,, —Adulteration of, 211. „ ,, with bar- ley, 211. —Fat of, 211. — Gliadin, 211. Oat-starch, 175, 177, 17S. ffinocyanin, 564. CEnolin, 564. Oidium aurantiacum, 202. Oil— Angelica, 494. ,, —Fixed, of mustard, 596. ,, of almonds, 617, 619. ,, of calamus, 493. ,, of cardamoms, 493. ,, of coriander, 494. ,, of hops, 512. ,, of juniper, 495. „ of mustard, 601. „ of olives, 623. ,, of pepper, 606. ,, ,, —Estimation of, 606. ,, of rice, 214. Oleic acid — Muter's process of deter- mining, 369. Olein, 240. ,, — Refraction of, 355. Oleo-margarine, 302-304. Oleo-refractometcr, 358. Oleum Theobromae, 451. Olive oil, 623-62S. ,, — Adulteration of, 625. „ — Chemical tests for, 625. ,, — Specific gravity of, 625, ,, — Spectrum of, 625. Olive stones as an adulterant of pepper, 610. Orange and yellow colours, 100-103. Oranges, 161. Orellin, 100. Orris root, 1/3. Osazones, 112, 126, 288. O'Sullivan and Valentine's analysis of malt, 508. Ox fat— Melting-point of, 354. Oxalate of methyl, 480. Oxazines and Thiazines, 110. Oxygen absorbed by milk, 252. ,, in water, 650. ,, process — Standard solution for determining, 693. Oxy ketone colours, 110. Pachyeuizus angulatus, 177. Pale ale, 505. Palladium solution, 693. Paimella Prodigiosa, 202. Palmitic acid, 239. Palmitin, 239. Panthaleon's views on milk, 230. Paraffin — Patterns of, 347. Paraphenylene blue, 111. violet, 111. Paraxylene as a solvent in crj'oscopie determinations, .375. Parmesan cheese, 380. PartheiFs estimation of gl j'cerin, 562. Parturient apoplexy — Milk in, 295. Pasteur's research on fermentation,. 499. Patterns of fats, 346. Pavy's method of sugar titration, 144. Pea cheese, 227. ,, starch, 178. „ soup, German, 223. ,, tablets, German, 22.3. Peach, 159. kernels, 618. Pear, 159. Pearl barley, 212. Peart v. Edwards, 60. Peas, 173, 178, 222-227. ,, — Poisoning by putrid pre- served, 228. Pepper, 603-614. ,, —Adulteration of, 609. „ —Analysis of, 609. —Ash of, 609. ,, — Bibliography relative to, 613. ,, — General composition of, 604. , , — Legal case relative to, 6 1 2. ,, — ^licroscopicalstructureof, 604. 730 Pepper — Nitrates of, 610. —Oil of, 606. ,, — Redi's experiments on, 33, ,, — Starch of, 176. ,, — Structures of, 603. ,, — Varieties of, 604. Pepper corns— Artificial, 612. Pepperette, 610. Pepperers or spicers — Regulations relative to, 10. Peppermint drops, 150. Permanganate alkaline, 693. ,, process, 656. , , of potash — Absorp- tion coefficient, 86. Peronospora infestans, '219. Perroucito— Case recorded by, 302. Pfefter-muuz liqueur, 497. Phaiophyll, lUo. Pharinacopa'ias — Establishment of, 11-18. Phenol test for nitrates, 63S. Phloroglucol as a test for olive stones, 611. Phloxine, 96, 112. Phospliates in the ash, 119. ,, in water, 636. Phosphiue, 103, 111, 114. Phosphomolybdic acid, 453. Phosphorus— Organic, in milk, 280. Photography, 89. Phthisical milk, 296, 300. Phjdlocyanic acid, 224. Phyllocyanine, 104. Phylloxanthine, 104. Phytelephas macrocarpa, 437. Phytochoiesterin, 281. Phytolacca, 577. Picini's method of estimating nit- rites, 637. Picric acid, 100, 112, 114, 509, 533, 538. Piorotoxin — Detection of, 509, 537. Piesse & Stansell's analysis of mus- tard, 595. Pigeon — Milk-like secretion of, 261. Piperic acid, 607, 609. Piperidin, 606. Piperin, 606, 609. Placitus' views on milk, 230. Plastering of wine, 578. Playfairs experiments on the feeding of cattle, 321. Pliny, 4, Plum, 159, 100. Plum kernels— Content of amyg- daliu, 018. Poirier's yellow, 114. Pneumonia — Milk in, 112. Poivrette, 610. Polarimetric estimation of sugar, 145. Polarisation of wine, 552. Police des commissionaires, 12. Poison's patent flour, 210. Ponceau, 114. 3R.,96. 2R.,96. B. & R., 566. Potassium iodide solution, 093. , , monochromate — Solution of, 693. ,, ])ermanj:anate — Solution of, 693. Pope V. Turle, 50. Poppj' oil — Constants of, 359, 624. —Test for, 62S. Port — Composition of old, 546. Porter — Composition of, &c., 504. Portugal berries as a colouring of wine, 570, 573, 574. Potato, 218. ,, — Analysis of, 220. fungus, 219. ,, starch, 172, 176, 178. ,, ,, as an adulterant of wheat flour, 190. Power, Dr. Hy., 30. Prejudice question, 29-31, 700. Prinmline base, 113. Preusse and Tiemann's estimation of oxygen in water, 648. Private purchase of samples, 54. Privet as a colouring of wine, 571, 573, 575, 577. Proof spirit, 467. Prosecution of offenders against Sale of Food and Drugs Act, 59. Prune — Composition of, 159. Prunier's method of separating coffee from chicory, 440. Public analyst— Appointment of, 52. Public Analj'sts' Society, 42. Purchase of samples, 53, 56, 62, 702. Purple cow wheat, ISO. Purpurine, 98. ,, as a test for alum, 195. Pyronine, 111. QuASSiiK, 517, 534. Quercetiu, 401. 731 Quercitannic acid, 401. Quercitron, 116. Quinoline colours, 110. Rabourdin, H., metliod of analysing pepper, 611. ,, ), of estimating pepper, 611. Railwaj^ Station — Taking of samples at, 6;^. Raspberry, 99, 159, 162. Rectified spirit, 467. Red bread fungus, 202. Red colouring matter, 92-99. Redi, Francesco, .32. Reduced extract of wine, 560. Refractometers, 355. Regnault's volumemometer, 433. Refusal to sell, 50, 55. Reichert's process for saponifying butter, .357, 365, 366. Reinhardt's method of taking melting points, 351. Reports — Quarterly, of analyst, 702. Resorciu as a test for nitrates, 538. ., ,, sugar, 286. blue, 112. ,, yellow, 177. Rhinanthin, 187. Rhizopus nigricans, 202. Rice, 167, 177, 214. ,, starch, 175, 178. ,, vinegar, 567. Riddle's patent process for the pre- servation of milk, 319. Ridge's food, 209. Riegler's method of estimating alco- hol and extract in wine, 547. Ritthausen's copper process for the analysis of milk, 277. Robine's method for detecting potato starch in wheat flour, 191. ,, ,, for estimating the value of flour, 197. Rocceline, 114, 566. Rodriguez's method of examining flour, 193. Rook V. Hoi^ley, 61. Roquefort cheese, 380, 382. ,, ,, — Ripening of, 3S1. Rosaniline, 93, 566. Rose Bengal, 112. Rosocyanin, 100. Rosolic acid, 96. Rouch V. Hall, 55. Rum, 490. P^ussell and Laplace, spectrum of water, 631. Rutic acid, 242. Rye, 213. ,, bread, 214. ,, —Digestibility of, 201 „ —Fat of, 213. ,, flour, 213. ,, starch, 174, 176. SACCHARiMiiTRE a Penombrcs, 150. Saccharin, 103. Sacchse, A. — Solution for the esti- mation of sugar, 142. Safflower, 115. Saffron — Old penalties for adulterat- ing, 16. Safranine, 93, 111, 114, 566. Saglier's method of testing alcohol, 4S5. Sago in pepper, 612. ,, starch, 174, 177. Saladin of Ascala, 32. Sale of Food and Drugs Act, 47. Text of, 699. ,, ,, ,, Amendment Act— Text of, 708. Salicin in beer, 536. Salicylic acid in beer, 540. in milk, 316. Salkowski's test for cholesterin, 2S1. Salt in beer, 540-543. Sample — Division of, into parts, 58, 701. ,, — Ti-ansmission through the post, 702. Sand in pe])per, 612. Sande, J. B. Vanden, 33. Sandys v. Small, 52. Santalin, 99, 115. Saprophiles, 600. Saponin, ISS. Savalle's method of testing alcohol, 484. Scala's analysis of rum, 491. Scarlatina in relation to milk, 298. Scheele — Discovery of lactic acid, 236. Scheibler's method of detecting dex- trin in sugar, 130. 732 Scheibler's method of determining sugar ash, 134. Schmidt and Hiepe's method of esti- mating acids in wine, 553. Schmidt's analysis of old wines, 550. „ process of detecting picro- toxine in beer, 537. ,, sj of estimating sul- phur dioxide in wine, 555. Schmoeger's observations on the polarisation of milk sugar, 284. Schoepff — Experiments on milk, 235. Schultz and Julius' scheme for colours, 109-117. Schiitzenberger on the decomposition of casein by baryta, 243. " Search after claret," 10. Sechium starch, 177. Section cutting, 73, 74. Sedgwick -Rafter method, 672. SelivanofFs method of detecting sugars, 286. Sell's process of estimating amyl alcohol, 484. Separated milk, 331. Sesame oil, 359, 624. Sewage farm — Influence of, on the milk, 325. Seyler-Hoppe views on casein, 242. Sheep's milk, 259. Sieben's analysis of hone}', 155. Silver nitrate— Standard solution of, 693. Sinalbin, 597. Sinapin, 596. Skalweit's refraction figures for butter, 355, 359. Skeleton ashes, 403. Skim cheese, 380. ,, milk, 331. Sloe, 409. Smec's views as to milk of cows on sewage farms, 325. Smith, Angus — Method of water- testing, 620. Smith, Dr. L.— Microscope, 72. Soap— Standard solution of, 693. Sodic nitrite— Solution of, 694. Sodium chloride— Solution of, 694. ,, hydrate— Solution for esti- mation of nitrates, 694. 5, hyposulphite solution, 694. ,, silicate as an adulterant of butter, 345. Soldaini's solution for estimation of sugar, 137. Soleil's saccharimeter, 147. Somerset House^Reference to, 60. Sorbinazone, 127. Sorby-Browning micro-spectroscope, 74. Sorby's investigation of the colouring matters of malt, 508. ,, method of detecting bitters in beer, 539. ,, method of measuring absorp- tion bauds, 76. Sow's milk, 260. Soxhlet's apparatus for ether, 37. , , experiments on the estima- tion of sugar by Fehling's solution, 139, 142. ,, method for obtaining dex- trose, 136. „ methodof detectingnitrates in milk, 292. ,, process of milk analysis, 274. Spanish chocolate, 450. Specific gravity — Determination of, by Boyle, 33. ,, of butter fat, 357, 359. ,, tables for alcohol, 46S-470._ Spectroscopic determination of metals, 79, 80. ,, examination of colour- in ir-matteis, 91. Spectroscopy — Quantitative, 84. Spectrum of water, 631. ,, analysis as applied to beer, 539. ,, wine, 569. Spermaceti — Pattern of, 346. Spirit indication (tables , 526-527. ,, vinegar, 582, 586. Spirits— Standard of strength, 49, 70S. Sponges in water, 677. Standard solutions in use in water analysis, 6'.l2. Starch, 163-178. ,, — Estimation of, by conver- sion into sugar, 164-169. ,, —Solution of, 694. Starches— Microscopical identifica- tion of, 170-17S. ,, — Quantitati\e determina- tion of, 171. 733 Stearin, 240. ttein's method for the detection of colouring matters, 114-117. ,, photographic camera, 90. Stephens' patent relative to milk pre- servation, 318. Stilton cheese, 379. Stock's method of detecting beef stearin in lard, 393. Stone's study of starch determina- tion, 1G9. Stout — Composition of, 505. Strawberries, 159, 161. Strychnine in beer, 53j. Stutzer's classification of nitrogenous constituents of cocoa, 463. Sublimation of theine, 397, 413. Succinic acid— p]stimation of, 561. Sudan, II. and III., 113. ,, brown, 113. Sugar, 126-150. ,, —Adulteration of, 128. ,, analysis, 132. ,, as an adulterant of beer, 543. —Ash of, 134. ,, candy, 151. cane in milk, 281-286. ,, compounds with bases, 126. ,, concretes (analj'sis of), 134. ,, cyanide process, of estima- tion, 143. ,, — Estimation of, 137-150. ,, ,, in wine, 563. ,, ■ — Full analysis of, 130. ,, Invert, 127, 139. Milk, 114. ,, — Pavy's process of estima- tion, 144. ,, test for sewage in water, 684. „ —Solubility of, 128. „ — Starch or glucose, detection of, in cane sugar, 129. Sullivan's method of starch estima- tion, 165-167, 181. Sulphates in milk, 246. ,, in water, 650. Sulpho-cj'anides in milk, 247. Sulphonated azines, 112. Sulphur in mustard— Estimation of, 601. Sulphuric acid in vinegar — Estima- tion of, 584, 585. ,, method of saponifi- cation, 367. Sulphurous acid — Estimation of, ia wine, 535. Sweetmeats, 151. SwissCompany'scondensedmilk,317. Tabarie's method of alcohol estima- tion, 477. Table beer, 505. Tacca arrowroot, 175, 178. Tallow — Melting-point of, 354. ,, — Patterns of, 347. Tannin — Determination of, 415, 416. ,, in brandy, 490. ,, in tea, 4i7. Tapioca, 175. Tartaric acid— Estimation of, in wine, 553, 562. Tea, 395-424. ,, — Acts relative to, 21. ,, — Adulteration of, 423. „ —Analysis of, 416-423. ,, — Bohemian, 423. ,, —Brick, 395. ,, — Chemical composition of, 397, 401. ,, — Chemical method of examin- ing leaves of, 396. ,, — Examination of, by Excise, 705. ,, — Facing of, 422. ,, leaf— Structure of, 396. ,, — Microscopical methods of de- tecting adulteration, 403. ,, — Varieties of, 395. Teinturier, a wine colouring matter, 573. Testi Ludovici on milk sugar, 231. Theine — Discovery of, by Leeuwen- hoek, 35. „ —Estimation of, 397, 412. ,, — Physical properties of, 397. ,, —Solubility of, 399. ,, —Tests for, 400. Theobromin, 452-455. ,, — Poisonous effects of, 454. Thiazines, 112. Thierry, M., Spectroscope of, 78. Thiobenzyl colours, 110. Thionine blue, 111. Thresh's method of estimating sul- phuric acid in vinegar, 586. Tin in peas, 225. Titer test for fats, 354. Tobacco — Ash of. 404. 734 Toffy, 151. Toluidine blue, 111. Toluylene red. 111. Tomlinson's cohesion figures, 3iS. Total solids of milk, 2(38. Tous les mois, 171. Traube's method of detecting fousel oil, 487. Treacle, 157. Tributyrin as au adulterant of butter, 3-43. Trifolium bitter, 534. Trinidad cocoa, 44S. Triiiitropheuol — See Picric acid. Tritisecaline, 179. Troiieoline, 114, 5GG. Tschirch's researches on copper, 224. Tubercular disease — Eelation of, to milk, 300. Turmeric, 100, 115, 173. ,, in mustard, 600. Twitchell's modification of Jean's process, 370. Typhoid bacillus — Characters of, 686. Typhus— Milk in, 298. Tyrotoxicon, 307, 385. Ulsch's methods of estimating nitric acid, 640, G42. Uraniue, 112. Urea in milk, 289. Urogleiia, G7S. Vacuum processes, 71. ,, for estimating the water in milk, 275. Valentine and 'Sullivan's analysis of malt, 508. Valenta's test for butter, 349. Vanilla chocolate, 450. Vaughan, Victor, on tyrotoxicon, 307. Veal-dripping — Melting point of, 354. Vegetable ivory in coffee, 437. Vernois and Becquerel's method of milk analysis, 267. Victoria blue, 114. ,, green. 111, 114. ,, yellow, 112. Vieth's analysis of cream, 329. ,, butter milk, 378. Villenin's experiments on tubercle in ^ milk, 300. Vinegar, 5S1-5S9. Vinegar — Adulteration of, 585. ,,' —Analysis of, 582, 586. ,, eels, 35. ,, malt, 582. ,, wine, 582. Violet acid, 112. ,, black, 112. ,, neutral, 111. Viscometry of butter, 360. Vitruvius, 4. Voelcker's analysis of goats' milk, 257. Vogel, classification of starches, 176. Vogel's method of delineating spec- tra, 76. ,, observations on purpurine and alum, 195. ,, test for ergot, 189. Volatile organic matter in water, 653. Volhard's method of estimating chlorine, 635. Voltelenus on milk, 234. VuUyanoz on milk, 234, Wallace on sugar ash, 132. Wallace's analyses of treacle, 158. Wanklyn and Eassie's patent for milk preservation, 319. Wanhlyn's metliod of estimating butyric acid, 371. , , method of milk analysis, 371. ,, • standards (water), 691. Warranty under the Sale of Food and Drugs Act, 704. Water analysis, standards of purity, 689. ,, — Biological examination of, 685. „ -Colour of, 630. ,, in brewing, 505, 506. ,, — Butter. 376. „ in milk, 277. ,, — Microscopical appearances of, 673. ,, — Mineral analysis of, 671. ,, organic! nitrogen estimated by Kjeldahl's method, 670. ,, — Quantitative analysis of, 633. „ —Smell of, 630. ,, — Total solid residue of, 634. Wave lengths, 72. '35 Weinwurm's analysis of nitrogenous constituents of cereals, 183. Weiske's experiments on milk in re- lation to feeding, 320. Weld, 116. Welman's colour test, 32. Wensleydale cheese, 3S0. Werner-Schmidt i^rccess of fat ex- traction, 275. Wheat flour, 179-198. ,, — Adulteration of, 186. ,, — Analysis of, 186. ,, —Constituents of, 179. ,, starch — Vogel's measure- ments of, J 76. Whey — Composition of, 385. Whisky, 491. Wiesuer's measurements of starches, 177, 178. Wigner's determinations of nitrogen in cereals, 183. ,, observations on milk fat, 239. Wiley's observations on carbohy- di-ates in milk fat, 28 1 , 282. Willard's observations on milk in relation to sewage, 325. Willow herb leaves as an adulterant of tea, 411. Wine, 545-580. ,, — Adulteration of, 546. ,, — Analysis of, 547, 579. ,, — Ancient adulteration of, 4. ,, — Ash of, 578. Wine — Bibliography of, 580. ,, — Changes of, through age,546. ,, astringent matters, 564. ,, — Colouring matters of, 564. ,, — Constituents of, 548-551. ,, —Contamination of, by lead, IS. ,, — Decree by Provost of Paris in 1871 relative to, 13. ,, — Electrolysis of, 565. ,, — Old English Act relative to, 20. ,, — Old German regvilations re- lative to, 16. ,, — Vinegar, 581. ,, — Watering of, 558. ^Vood^vard's method of photography, SO. ,, photometer, 90. Wormwood bitter, 535. Xanthophyll, 105. Yam starch, 176. Yeast — Chemical constituents of, 500. Yellow— Brilliant, 112. Zallonis' experiments on the inocu- lation of tubercle, 300. Zinc in water, 632. ., iodide starch, 632. Zipperer's method of determining cocoa-red, 456. BELL AND BAIN. LIMITED, PRINTEKS, GLASGOW. jrotelRD Edition. In large 8vo. Cloth, with Tallies and Illustrations. Price 21s. POISONS: Their Effects and Delection. By A. WYNTER BLYTH, M.R.C.S., F.C.S., Barrister-at-La\v, Public Analyst for the County of JJeyon, &c. "Undoubtedly the most complete \voiik on Toxicology. in. our langi^age."— 77i« Anahist (mi the Third Edition). '■ - . ■ ' "" " As a PR.4CTICAL GUIDE, we know no better work."— r/ic Lancet (on the Third Editiwi). HYGIENE (A Handbook of). By Surgeon-Major A..:M. . Davies, D.P.H. Carub., late Assistant-Professor of Hygiene, Army Medical School. Pocket . .Size. Leather. "With Illustrations. 12s. 6d. " This ADMIRABLE HANDBOOK."— .BtZiw. Med. Journal. " The elegant dress cf the little volume before us is but the outer coyering of a. TRULY RICH KERNEL, and justly merits the praise it spontaneously calls fortji. . . . Compact, handy, comprehensh^b, it certainly niciits a high place among the text- books of the Aa.y." Sanitary Record. DISINFECTION and DISINFECTANTS. With an Account of the Chemical Sulistances used as Antiseptics and Pkeseuvatives. By Samuel Kideal, B.Sc. (Lond.), F.I.C., E.C.S., Examiner in Chemistry to the Koyal College of Physi- cians. With Diagrams of the most approved Modern Appliances. 12s. 6d. " Dr. RiDEAL's volume is bound to prove of GREAT value."— P/iarmiaeewt.' Journal. " An exhaustive Treatise dealing with the whole kanoie of the subject :— Disin- fection by Heat, Chemical Disinfectants, Practical Methods, Personal Disinfection, Legal Regulations, and Methods of Analysis ... so very well done and so useful .that it will be valued by ALL connected with Sanitation and Public Health."— C/iemist and Druggist. PRACTICAL SANITATION. By George IIeid, M.D., D.P.H., Medi- cal Officer, Staffs. County Council. A Handbook for the Use of Medical Officers of Health and Others. With an Appendix on Sanitary Law by H. Manley, M.A., D.P.H., Medical Officer, WestBromwich. Third Edition. Illustrated, 6s. " A VERY useful handbook with a very useful appendix." — Sanitary Record. PRACTICAL PHARMACY and DISPENSING (Elements of). By Wm. Elborne, E.C.S., F.L.S., B.A. Cantab., Demonstrator of Materia Medica and Teacher of Pharmacy at University College, London, Pharmacist to University College Hospital. With Facsimiles of Prescriptions and Illustrations. Cloth, 8s. Cd. " Admirably adapted for Students of 'Sledicine."— Edinburgh Med. Journal. INORGANIC CHEMISTRY. By A. Duprb, Ph.D., F.R.S., and Wilson Hake, Ph.D., F.I.C., F.C.S., of the Westminster Hospital Medical School. Second Edition. 7s. (jd. " By a long way the best of the small manuals for Students."— Analyst. PETROLEUM AND ITS PRODUCTS. A Practical Treatise on the Geographical and (ieolot;ical (XcuzTence. clieniistry. Production, Refining, and Uses of Petroleum and Ozokerite; the Tehtiiiu, Traiisiiort, and Storage of Petroleum Products, and the Legislative Enaetnieiits lelating thereto ; together with a Descrip- tion of the Shale Oil Industry. By Buveutun Redwood, F.R.S.E., F I.C, Assoc. Inst. C.E., Consulting Adviser to the Corporaticm of London under the Petroleum Acts, &c. Assisted by CEii. T. HoLLOWAY, F I.C, A.R.C'.S., and ^Numerous Con- tributors. With Tables, -Maps, Plates, and Illustrations in the Text. In Two vols. Large 8vo, Handsome Chjtli. Price 45s. " Jlr. Redwood has produced a work which must be admitted TO THE rank of the 'indispensables' — a verdict rarely registered in these colunnis." — Journ.ofGasLightitig. OILS, FATS, WAXES, and ALLIED MATERIALS, and the Manu- facture therefrom of Candles, Soaps, and other products. By C. R. Alder Wright, D.Sc., F.R.S. Large Svo. With numerous Illustrations. 28s. " Dr. Wright's work will be found absolutely indispensable. Teems with in forfnation valualile alike to the Analyst and the Technical Chemist." — The Analyst. LONDON: CHARLES GRIFFIN & CO., LIMITED; EXETKU STREET, STRAND. ^^.-.. -c g^*^!?^: