COLUMBIA LIBRARIES OFFSITE HEALTH SCIENCES STANDARD HX64073491 R A425 H23 1 902 A manual of practica RECAP f^i?^\-l'y.i^r,?C:^-.XiC,yr'ti.:':^^ intl)f Cttpoflrmgork Co^y College of ^fjj»s!ician£i anb ^urseong From the Library of PROFESSOR PHILIP HANSON HISS 1868-1913 Donated by Mrs. Philip Hanson Hiss A MANUAL PRACTICAL HYGIENE STUDENTS, PHYSICIANS, AND MEDICAL OFFICERS. BY CHARLES HARRINGTON, M.D, Assistant Pkofessoe of Hygiene in the Medical School of Hakvabd L'niyeesity. SECOND EDITION, REVISED AND ENLARGED. ILLUSTRATED WITH TWELVE PLATES IN COLORS AND MONOCHROME, AND ONE HUNDRED AND THIRTEEN ENGRAVINGS. LEA BROTHERS & CO., PHILADELPHIA AND NEW YORK. Entered according to Act of Congress, in the year 1902, by LEA BROTHERS & CO., In the Office of the Librarian of Congress. All rights reserved. CoVY i WESTOOTT i. THOMSON. ELECTHOTYPERS, PHILADA. PREFACE TO THE SECOND EDITION. The demand for a second edition of this work within but little more than a year from the appearance of the first is gratifying evidence that the book in some degree has filled a reqmrement. During the interval, research in the field of hygiene has been active and fruitful, and the results have been incorporated in great measure in the present issue. A chapter on the relation of insects to human diseases, con- taining eight additional illustrations, has been added, parts of other chapters have been entirely rewritten, and throughout the book nu- merous changes and additions have been made. The amount of new material occupies about seventy pages, but the size of the book has been kept within reasonable limits by the excision of about thirty pages dealing chiefly with quarantine law and obsolete matter. The author entertains the hope that the second edition will l^e no less cordially received than the first. C. H. 688 BoYxsTON Street, Boston, August, 1902. Digitized by tine Internet Arciiive in 2010 witii funding from Open Knowledge Commons http://www.archive.org/details/manualofpractica1902harr PREFACE TO THE FIRST EDITIOI^. Ix preparing this work, the object of. the author has been to provide a students' text-book which should cover the most important topics included in the wide domain of Hygiene, and be useful in the laboratorv and as a reference book for practitioners and health officers. The subject is so broad that it is impossible to treat it in its entirety in a thorough manner in a single volume ; therefore, certain topics which find a place in some of the larger works, and which are of interest chiefly to a somewhat limited circle of specialists, have not been considered. Of late, it has become the custom to incorporate in works on liygiene a chapter on elementary bacteriology. It has seemed to the author that, inasmtich as a knowledge of this closely allied science is recognized as a very essential part of the equipment of the modern medical practitioner, and is taught, as its importance deserves, either as a separate subject or in connection with pathology, it would be as much a work of supererogation to give a brief description of species and technic herein as to include a chapter on elementary chemistry, physics, or other affiliated great subject. It is assumed that the reader has already accj[uired at least a fair working knowledge of bacteriology, or that, lacking it, he will turn rather to special works in which the science is fully treated. Again, certain topics, which not infrequently are included in works of this nature, but which lie more properly within the fields of engi- neering and architecture, such, for example, as the construction of aqueducts and sewers, the nature and strength of building materials, and the arrangement of hospitals, have been deemed as hardly within the province of the hygienist, and, consequently, have been excluded. To those who, by their writings and otherwise, have been of assist- ance in the preparation of this book, the author extends his sincere thanks ; and especially to his friend and colleague Professor Harold C. Ernst for taking the photomicrographs from which the plates illus- trative of the starches and trichinae have been prepared. C. H. 688 BoYLSTOx Street, Boston. 5 CONTENTS. CHAPTER I. PAGE FOODS 17 § 1. GeSTEEAX, CojrSIDEIlATIO^'S 17 The nutritive value of foods, 17. Amount of food necessary, 18. Composi- tion of foods, 19. Proteids, 19. Fats, 21. Carbohydrates, 21. Organic acids, 22. Inorganic sahs, 23. § 2. AN-TAf A T, Foods : Meat?, Fish, Eggs, axt> Meat Peodtjcts 23 Meats, 2-1. Digestibility, 24. Flavor, 24. Texture, 25. Effects of cooking, 25. Characteristics of good meat, 26. Comparative digestibility of meats, 26. Eed meat and white meat, 27. Composition of meats, 27. Beef, 28. Pork, 29. Yeal, 30. Mutton, 30. Lamb, 30. Poultry, 31. Horse meat, 31. Meat preparations, 32. Sausages, 32. Fish, 34. Digestibility, 35. Keeping qualities, 35. Composition, 35. Meat and fish, and pai-asitic disease, 37. Transmission of disease by meat and fish, 42. Tuberculosis, 43. Typhoid fever and cholera, 48. Poisoning by meat and fish, 53. Poisoning due to substances normally present in the living organism, 53. Poisoning due to bacterial products in meats and fish, 53. Onset and coui^e of s\Tnptoms, 57. Xature of symptoms, 57. Post-mortem appearances, 58. Character of meats which cause poisoning, 58. Cases illustrative of poisoning by fish and meats, 60. Poisoning by mussels, 60. Poisoning by henings, 61. Poisoning by salmon, 63. Poisoning by oysters, 63. Poisoning by veal, 64. Poisoning by pork, 66. Poisoning by beef, 70. Poisoning by hoi-se meat, 73. Poisoning by sausages, 74. Poisoning by kid meat, 76. Meat inspection and slaughter- ing, 76. Eggs, 78. Lard, 81. § 3. iJjXK -Ajs'D Milk Peodccts . 82 Composition of cows' milk, 82. Fat, 83. Milk sugar, 84. Proteids, 84. ^Mineral matter, 85. Specific gi-avity, 85. Eeaction, 85. Appearance, 85. Taste, 86. Presence of alcohol, 86. Colostrum, 87. Changes produced in milk by boiling, 87. Changes due to bacterial action, 88. Preservation of milk, 91. Adulteration of milk, 94. Condensed milk, 95. Koumiss and kefir, 95. Cream. 96. Milk as a factor in the spread of disease, 96. Poison- ous mUk, 96. ;Milk fi-om diseased cows, 98. Milk contaminated from with- out with organisms related to human diseases, 104. Diphtheria, 105. Cholera, 105. Scarlet fever, 106. T%-phoid fever, 107. Cholera infantum, 108. Analysis of milk, 110. DeteiTnination of specific gravity, 111. Determination of fat, 112. Determination of total solids, 115. Determina- tion of milk sugar, 116. Detennination of ash, 117. Determination of pro- teids, 117. Detection of added coloring matters, 118. Detection of preserva- 7 8 COyTENTS. PAGE tives, 120. Methods of distinguishing between raw and cooked milk, 122. Detection of gelatin in cream, 123. Butter, 124. Butter as a carrier of dis- ease, 127. Analysis of butter, 128. Cheese, 132. Composition of cheese, 134. Adulteration of cheese, 135. Analysis of cheese, 135. Cheese as a cause of poisoning, 136. § 4. Vegetable Foods 136 Farinaceous seeds, 137. Cereals, 137. Wheat, 137. Composition of wheat, 138. Wheat flour, 139. Preparations of wheat flour, bread, 140. Macaroni, spaghetti, and vemiicelli, 142. Adulteration of flour, 143. Eye, 144. Barley, 144. Oats,. 145. Corn, 146. Eice, 147. Buckwheat, 148. Legumes, 148. Peas, 149. Beans, 150. Lentils, 151. Farinaceous preparations, 151. Sago, 151. Tapioca, 151. Arrowroot, 152. Fatty seeds (Nuts), 152. Almonds, 152. Cocoanuts, 152. Walnuts, 152. Peanuts, 153. Chestnuts, 153. Vegetable fats, 154. Olive oil, 154. Cotton-seed oil, 154. Tubei-s and roots, 154. Potatoes, 154. Sweet potatoes, 157. Artichokes, 157. Eoots, 157. Herbaceous articles, 158. Fruit products used as vegetables, 158. Fruits, 159. Apples, 159. Peal's, 160. Peaches, 160. Apricots, 160. Plums, 160. Cherries, 160. Oranges, 160. Grapes, 161. Melons, 161. Bananas, 162. Figs, 162. Berries, 162. Edible fungi, 163. Mushrooms, 163. Trufiies, 163. Saccharine j)reparations, 163. Cane sugar, 163. Maple sugar, 164. Glucose, dextrose, 164. Molasses, 165. Honey, 165. Confec- tionery, 167. Jellies and jams, 168. § 5. Beveragks 168 Stinuilant beverages containing alkaloids, 168. Tea, 168. Adulteration of tea, 170. Coffee, 171. Cocoa, 173. Fermented alcoholic beverages, 175. Beer, 175. Process of manufacture of beer, 177. Substitutes for barley malt, 178. Substitutes for hops, 179. Physical properties and chemical composi- tion of beer, 180. Adulteration of beer, 180. Analysis of beer, 181. Tables showing percentage of alcohol by weight and by volume, 182. Detection of preservatives, 188. Wines, 189. Classification of wines, 191. Composition of wines, 191. Adulteration of wines, 192. Analysis of M'ines, 194. Detec- tion of preservatives, 195. Cider, 196. Perry, 197. Distilled alcoholic beverages, 197. Brandy, 198. Whiskey, 199. Eum, 200. Gin, 201. Liqueurs, 201. § 6. Condiments, Spices, and Bakers' Chemicals 202 Vinegar, 202. Lemon juice and lime juice, 204. Salt, 204. Mustard, 205. Pepper, 205. Cloves, 205. Cinnamon and cassia, 206. Allspice, 206. Ginger, 206. Nutmeg, 206. Mace, 206. Cayenne pepper, 206. Baking powders, 206. § 7. Food Preservation 208 Cold, 209. Drying, 209. Salting, 209. Smoking, 209. Canning, 209. Chemical treatment, 211. Boric acid and borax, 213. Salicylic acid, 216. Sulphites, 217. Formaldehyde, 217. Hydrogen peroxide, 218. Sodium fluoride, 219. Sodium bicarbonate, 219. § 8. Contamination of Foods by Metals 219 Copper, 219. Lead, 221. Zinc, 222. Nickel, 222. Tin, 222. Metallic con- tamination from kitchen utensils, 222. CONTENTS. ■ 9 CHAPTER II. PAGE AIR 224 Oxygen, 224. Nitrogen, 226. Argon, 227. Hydrogen, 227. Carbon dioxide, 227. Ozone, 229. Peroxide of hydrogen, 230. Ammonia, 230. Nitrogen acids, 230. Aqueous vapor, 231. Dust and micro-organisms, 233. Carbon monoxide, 235. " Sewer gas," 237. Organic matters, 240. EfTects of vitiated air, 240. The air as a carrier of infection, 244. Influence of fog, 251. Examination of air, 251. Determination of aqueous vapor, 252. Deter- mination of relative humidity, 253. Glaishei-'s table, 254. Table of tensions, 255. Determination of carbon dioxide, 257. Determination of carbon monoxide, 265. Determination of ozone, 266. Determination of dust, 267. Bacteriological examination, 268. CHAPTER III. THE SOIL 270 The soil, 270. Constituents of the soil, 272. Physical properties of soils, 273. Pore-volume, 273. Permeability of soils, 274. Capacity for water, and water-retaining capacity, 278. Soil temperature, 280. Changes in the char- acter of soils due to chemical and biological agencies, 282. Soil-air, 283. Soil-water, 286. Sources of soil-water, 289. Loss of soil moisture by evapora- tion, 289. Influence of vegetation on soil moisture, 290. Other eflects of vegetation upon the soil, 291. Pollution of the soil, 292. Bacteria of the soil, 294. Soil and disease, 296. Soil dampness and disease in general, 296. Soil and j)ulmonary tuberculosis, 297. Typhoid fever, 297. Cholera, 300. Bubonic plague, 300. Diphtheiia, 301. Malaria, 302. Tetanus and malignant «dema, 303. Anthrax, 304. Epidemic diarrhoea, 305. Goitre, 306. Yellow fever, 306. Other diseases, 306. Examination of soils, 307. Pore-volmue, 308. Permeability to air, 309. Permeability to water, 310. Water capacity, 312. Capillarity, 312. Moisture, 312. Organic and volatile matters, 313. Deter- mination of CO2 in soil-air, 313. Bacteriological examination of soil, 315. CHAPTER IV. WATER 316 Introductory, 316. Rain, 316. Surface-waters, 317. Ground- waters, 318. Physical and chemical characteristics of water, 320. Appearance, 321. Re- action, 322. Odor, 322. Substances found normally in water, 324. Gases, 324. Carbon dioxide, 325. Organic matter, 325. Ammonia, 326. Albumi- noid ammonia, 328. Nitrites and nitrates, 328. Mineral matters, 330. Salt, 330. Hardness, 331. Bacteria in water, 332. Water supplies, 335. Stored rain, 335. Surface-waters, 337. Ground-waters, 338. Springs, 338. Wells, 339. Driven wells, 340. Drainage area of wells, 344. Filter galleries, 346. Classification of waters from the sanitary standpoint, 347. Purification of water, 349. Oxidation, 349. Dilution, 350. Sedimentation, 350. Bacterial action, 350. Vegetation, 350. Methods of purification, 351 . Chemical treat- ment, 351. Boiling and distillation, 355. Filtration, 355. Domestic filters, 355. Filtration of public supplies, 357. " Mechanical filtration," 363. Plaque filters, 363. Removal of hardness, 364. Removal of iron, 364. Action of water on lead and other metals, 365. Action on iron, 369. Action on zinc, 10 CONTEXTS. PAGE 369. Action on tin, 370. "Water and disease, 371. Disordei's connected with mineral matter, 372. Disordei-s connected with organic pollution, 374. Ty- phoid infection of water supplies, 376. Influence of introduction of public water supplies on typhoid rates, 377. Classification of cities according to typhoid fever death-rates, 379. Examples of typhoid fever epidemics and of limited outbreaks traced to infected water, 381. Asiatic cholera, 386. The propagation of cholera in India, 388. Parasites and drinking-water, 391. Ice, 393. Chemical examination of water, 394. Collection of samjiles, 394. Determination of free and albuminoid anmionia, 395. Determination of other nitrogen compounds, 400. Nitrites, 400. Nitrates, 401. Determination of chlorine, 402. Determination of residue, 403. Determination of hardness, 403. Determination of "Oxygen required," 404. Detemiination of color, 405. Determination of odor, 405. Determination of reaction, 405. Deter- mination of turbidity, 406. Detection and detei-mination of lead, 406. Detection of tin, 409. Detection and detemiination of iron, 409. Inferences as to character of water from the results of sanitaiy chemical analysis, 409. Bacteriological examination of water, 412. Collection of samples, 413. Plant- ing of samples, 413. Quantitative determination, 414. Qualitative deter- mination, 415. Comparative value of chemical and bacteriological analy.sis of water, 417. CHAPTER V. IL\BITATIONS, SCHOOLS, ETC 420 § 1. General Con.^i derations 420 Aspect, 420. Construction and arrangement, 420. Care of habitations, 421. vSchools, 421. School fimiiture, 422. Legislation concerning schools, 423. § 2. Ventilation and Heating 423 Ventilation and heating, 423. Amount of space required for good ventilation, 425. Natural forces in ventilation, 426. Diffusion and gravity, 427. Peri- flation and aspiration, 429. Natural ventilation, 431. Inlets and outlets, 432. Mechanical ventilation, 435. Artificial heating in its relation to ventilation, 435. Radiation, 436. Conduction, 436. Convection, 436. Methods of warming, 437. Open fires, 437. Stoves, 437. Furnaces, 439. Hot-water pipes, 439. Steam pipes, 439. Eegulation of temperature, 440. Necessity of providing moisture. 441. Detemiination of rates of ventilation, 443. § 3. Lightino -i^-'> Natural lighting, 445. Artificial lighting, 447. Luminosity of flame, 447. Gas burners, 448. Varieties of illuminating gas, 449. Coal-gas, 449. Water- gas, 449. Acetylene gas, 450. Gasolene gas, 451. Ini[)Uiities given off" in lighting, 451. Gas pipes, 451. Fixtures, 452. Electric lighting, 452. § 4. Pltmring -i-^'^ Plumbing, 452. The soil-pipe and main drain, 454. "Waste pipes, 459. Traps, 460. Grease traps, 464. Loss of seal, 466. Non-siphoning traps, 468. "Water- closets, 470. The pan closet, 473. The plunger closet, 474. Hojiper closets, 475. Open wash-out closets, 475. Siphon closets, 476. Flushing apparatus, 478. "Water-closet connections, 479. I'rinals, 480. "Wash basins, 480. Bath- tubs, 482. Sinks, 484. Laundry tubs, 485. House service tanks, 485. Service pipes, 486. Testing plumbing, 487. CONTENTS. 11 CHAPTER VI. PAGE DISPOSAL OF SEWAGE 488 Sewage, 488. Methods of sewage disposal, 491. Discharge into the sea, 491. The pail system, 492. Chemical treatment, 493. Action of sewage effluents on fish life, 494. Sewage irrigation, 494. Influence of sewage irrigation on health, 498. The Waring system of irrigation, 499. Sewage filtration, 500. Other biological processes, 502. Dibdin's bacteria filter, 502. Cameron's septic tank, 504. The Scott-MoncriefF system, 505. CHAPTER VII. DISPOSAL OF GARBAGE 507 Grarbage, 507. Methods of disposal of garbage, 508. Incineration, 508. Ke- duction, 509. CHAPTER VIII. DISINFECTANTS AND DISINFECTION 511 Disinfectants, 511. Physical agents, 511. Light, 511. Heat, 514. Dry heat, 515. Steam, 515. Boiling water, 519. Cold, 519. Chemical agents, 520. Oxygen, 522. Ozone, 522. Hydrogen peroxide, 523. Chlorine, 524. " Chloride of lime," 524. Sodium hypochlorite, 526. Bromine and iodine, 526. Sul- phur dioxide, 526. Lime, 527. Ferrous sulphate, 528. Ferric sulphate, 529. Ferric chloride, 529. Zinc chloride, 529. Aluminum chloride, 529. Potassium permanganate, 530. Copper sulphate, 530. Mercuric chloride, 530. Mineral acids, 532. Organic substances, 532. Carbolic acid, phenol, phenic acid, 532. Cresols, 534. Creolin, 535. Lysol, 536. Bacillol, 536. Lysoform, 536. Saprol, 536. Solveol, 536. Solutol, 537. Alcohol, 537. Essential oils, 538. Soaps, 538. Medicated soaps, 541. FoiToaldehyde, 542. Formaldehyde apparatus, 543. Germicidal properties, 549. Power of pene- ti-ation, 552. Conditions favoring action, 553. Toxicity, 554. Amount neces- saiy for room disinfection, 556. Disadvantages, 556. Other applications of foi-maldehyde, 557. Prevention of dissemination of infectious material: Practical disinfection, 557. Disinfection of faeces, 558. Urine, 559. Sputum, 559. Discharges from the mouth, etc., 559. Eating utensils, 559. Bed linen and clothing, 559. Hands, 560. Air, 560. Eoom disinfection, 561. Disiij- fection of books, 564. Disinfection of water-closets, 564. CHAPTER IX. MILITAEY HYGIENE 565 Introductory, 565. The recruit, 567. Age, 568. Height, 571. Weight, 572. Examination of the recruit, 572. Chest capacity, 573. Grounds for rejection, 573. Hygiene of the soldier, 575. Personal cleanliness, 575. Contentment and cheerfulness, 575. Clothing of the soldier, 576. Wool, 576. Cotton and linen, 576. Shoddy, 576. Color, 577. Militaiy dress coats, 577. Gaiters and leggings, 577. Head covering, 577. Stockings, 578. Boots, 578. Under- clothing, 579. Abdominal bands, 579. Waterproof blankets, 580. The sol- diei-'s exercise and work, 580. Marching, 580. Care of the feet on the march, 585. Care of other parts, 586. The soldier's food ; " rations," 587. Alcohol 12 CONTENTS. PAGE in the ration, 589. Preparation of food, 591. Is the U. S. ration adequate m amount and composition? 591. Is the U. S. ration suited to the tropics? 593. Tropical dietaries, 595. Posts and camps, 597. Barracks, 599. Tents. 601. Water supply, 604. Sewerage, 604. Sinks and latrines, 605. Inspections, 606. Sanitary police, 607. The diseases of the soldier, 613. Typhoid fever, 614. Dysentery, 615. Malaria, 615. Measles, 616. Diarrhoeal diseases in general, 616. Sunstroke, 616. Venereal diseases, 617. CHAPTER X. NAVAL AND MARINE HYGIENE 618 Naval recruits, 618. The naval ration, 619. Water supply, 620. The sailor's sleeping rpiarters, 621. The diseases of the sailor, 622. Ventilation of vessels, 623. General hygiene of ships, 624. CHAPTER XI. TROPICAL HYGIENE 627 The soldier and the civilian in the tropics, 627. Habits of life, 629. Diet, 620.^ The use of alcohol in the tropics, 631. Clothing, 633. Care of the person, 634. Tropical disea.ses, 634. CHAPTER XII. THE RELATION OF INSECTS TO HUMAN DISEASES 636 Introductory, 636. Flies, 637. Flea.s, 640. Bedbugs, 640. Mosquitoes, 641. Mosquitoes and malaria, 642. The malarial parasite, 645. Preventive measures, 651. Mosquitoes and yellow fever, 652. The yellow fever mos- quito, 6.54. Preventive measures, 65(5. Mosquitoes and lilarial disease, 657. Mos(|uitoes and dengue, 659. Mosquitoes and distomiasis, 660. CHAPTER XIII. HYGIENE OF OCCUPATION 661 Introductory, 661. Classification of occupations, 666. Occupations involving exposure to air vitiated by resjjiration, 668. Occupations involving- exposure to irritating and poisonous g:ises and fumes, 668. Occupations involving exposure to irritating and poisonous dusts, 674. Occujjations involving exposure to infective matter in dust, 680. Occupations involving the inhalation of offensive gases and vapoi-s, 681. <)ccuj)ations involving exposure to extremes of heat, 682. Occnpatinns involvmg exposure to dampnes.s, 682. Occupations involving exposure to abnormal atmospheric pressure, 682. Occupations involving con.strained attitude, 683. Occupa- tions involving overexercise of parts of the body, 684. Occupations involv- ing sedentary life, 684. Prophylaxis in general, 684. Employment of women and children, 685. CHAPTER XIV. VITAL STATISTICS 686 Introductory, 686. The census, 687. Estimated population, 689. Increase of population, 690. Population constitution, 690. Registrars' returns, 691. CONTENTS. 13 PAGE Marriage-rates, 692. Birth-rates, 693. Death-rates, 695. Influence of sex, 695. Influence of age, 695. Influence of race, 696. Influence of density, 697. Weekly death-rates, etc., 698. Zymotic death-rate, 698. Infantile death-rate, 698. High and low death-rates, 700. Correction of death-rates, 701. Classification of causes of death, 702. Eegistration of sickness, 702. Duration of life, 703. Probable duration of life, 703. Mean duration of life, 703. Expectation of life, 703. Life tables, 704. CHAPTER XV. PERSONAL HYGIENE 706 § 1. Caee of the Person. Baths 706 § 2. Regulation of the Diet . 708 § 3. Rest and Recreation 708 § 4. Physical Exercise 709 Effects of active exercise, 709. Circulation and respii-ation, 709. Skin, 710. Nervous system, 710. Digestive apparatus, 711. Kidneys, 711. Efiect of exercise on weight, 711. Amount of exercise required, 712. Kinds of exer- cise, 713. Golf, 713. Wheeling, 714. Tennis, etc.. 714. Rowing, 714. § 5. Clothing 714 Color, 714. Texture, 714. Heat conductivity, 715. Hygroscopicity, 715. Materials, 715. Wool, 716. Silk, 717. Cotton, 718. Linen, 719. Rubber, 719. Leather, 720. Fur, 720. Felt, 720. Adulteration of clothing, 720. Chemical analysis, 721. Microscopical examination, 721. Poisonous dyes, 721. Selection of clothing, 722. CHAPTER XVI. VACCINATION AND OTHER PREVENTIVE INOCULATIONS 724 Vaccination, 724. Other preventive inoculations, 729. Asiatic cholera, 729. Bubonic plague, 729. Diphtheria, 731. Typhoid fever, 731. CHAPTER XVII. QUARANTINE 732 Quarantine, 732. Law of 1893, 734. Intei-state quarantine, 738. State quar- antine, 738. Sanitary cordon, 740. Municipal quai-antine, 740. Camps of detention, 741. CHAPTER XVIII. DISPOSAL OF THE DEAD 742 Earth burial, 742. Sites for cemeteries, 744. Cremation, 744. History of modem cremation, 746. PRACTICAL HYGIENE. CHAPTER I. FOODS. Section 1. GENERAL CONSIDERATIONS. Foods may be said to include everything taken into the system capable of being utilized directly or indirectly to build up normal structure, repair waste, or produce energy in any form, but in the common acceptation of the term they include only those substances which can be oxidized in the system, thus excluding water and air, without which the functions of the body can not be performed. Diet is a mixture of food materials of various kinds habitually taken in such quantity as is needed to maintain or improve the condition of the system. The Nutritive Value of Foods. The potential energy of food is measured by the amount of heat which can be obtained by its complete combustion, and is expressed in units of heat or calories. The amoimt of energy required to raise the temperature of 1 kilogram of water 1 ° C. is known as a large calorie ; that required to raise the temperature of 1 gram to the same extent is known as a small calorie ; thus, 1 large calorie equals 1000 small calories. When the term is used without qualification, large calories are understood. These heat units correspond to 425.5 units of work ; that is to say, the same amount of energy required to raise the tempera- ture of the given weight of water 1° C. is sufficient to raise 425.5 times the weight one meter. The amount of work done is expressed in kilogram-meters. The heat unit of the English system is the amount of energy required to raise the temperature of a pound of water 1° F., and its mechanical equivalent is 772 units of work ; that is to say, the same amount of energy will raise 772 pounds one foot. According to the English method, work done is expi'essed in foot tons rather than in foot pounds. The calorimetric values of different food materials express also their 18 FOODS. physiological values as nutriment. Rubner^ determined the calori- metric value of the proximate principles as follows : One gram of proteids = 4.1 calories. One gram of carbohydrates = 4.1 " One gi"am of fat = 9.3 " In the system, the carbohydrates and fats are burned completely to carbonic acid and water, but the proteids leave a residue of urea incapa- ble of still further oxidation. It is estimated that the physiological value of the proteids averages only about 75 per cent, of their calori- metric value. The calorimetric value of diifereut foods of the several classes is not to be understood as bemg exactly the same. Thus, two kinds of sugar or two kinds of fat, or two kinds of proteid have not exactly the same calorimetric values, and the figures above given are to be under- stood only as fair averages. Rubner^ has determined by actual ex- perimentation the relative calorimetric values of certain food materials, and has shown that 100 grams of fat are equivalent to 225 of syntonin, or 232 of starch, or 234 of cane sugar, or 243 of muscle fiber, or 256 of grape sugar. In other words, these several amounts of food material are isodynamic. Amount of Food Necessary. For the maintenance of a proper degree of health and strength, the individual must ingest an amount of food sufficient to meet the daily loss of nitrogen and carbon. This must necessarily vary according to circumstances, and hence no rule can be laid down to fit all cases. The best that can be done is to make certain general rules based on the amount of work ])erf()rmed, for the greater the amount of work done, the greater the amount of food required to meet the necessary consump- tion of fuel and to repair the tissues. When performing heavy labor, the naturally increased desire for food is shown particularly in the direction of fats, and secondarily of proteids. It has been estimated by Voit, after much experimentation and from voluminous data, that a man weighing 70 to 75 kilos (154 to 105 pounds) and working at moderately hard labor 9 to 10 hours a day requires 118 grams of proteids, 56 of fiit, and 500 of carbohydrates. This is equivalent to 3054.6 calories, and is regarded generally as the most accurate estimate obtainable. From Voit's and other figures, a number of writers have endeavored to construct standard dietiiries for the various conditions of bare subsistence, rest, and the performance of different amounts of daily labor, the subsistence diet being reckoned as sufficient only for the internal work of the body in the absolutely ]>assive condition, and that for rest as meeting the requirements of very gentle exercise. The constituents of the standard diets are prescribed in the follow- ing proportions : For each part of proteids, two-thirds of a part of fat, ' Lehrbuch der Hygiene, Leipsic, 1900, p. 438. * Ibidem, p. 430. COMPOSITION OF FOODS. 19 three and one-sixth parts of carbohydrates, and one-fourth of a part of mineral matter. The proportion of 1 part of nitrogen to 15 of carbon should be maintained as nearly as possible. In all dietaries it is necessary to have the different constituents properly proportioned, and in addition to vary from day to day the articles belonging to the same classes, since otherwise they will become distasteful. The question of nutritive value, while paramount, is not all, for taste and variety of flavor are important in a high degree, and the palate must be flattered. Knowing the composition of various food materials, the nutritive value of a given weight, and problems of supplying given amounts of proteids, fats, and carbohydrates from means at hand can be worked out mathematically. Composition of Foods. The constituents of food materials are partly organic and partly in- organic. The organic constituents include proteids, fats, carbohydrates, and organic acids ; the inorganic include water and mineral salts. Proteids. — The proteids are tlie most important constituents of both animal and vegetable foods, and their presence is necessary for the carrying on of all the phenomena of life. They are very complex colloid substances composed of carbon, oxygen, hydrogen, nitrogen,, and sulphur, possessing common properties and connected in very close genetic relationship. They are divided into animal proteids and vege- table proteids, but between the members of the one class and those of the other there are no important chemical differences, and they are about equal in nutritive value. Some of the vegetable proteids are not colloids, for according to Schmiedeberg, Weyl, Maschka, and others, they are found in crystalline form in the tissues of certain plants, and notably in the bean, pea, lentil, and various nuts. Proteids when completely split up by acids yield as end products ammonia, nitrogen, organic bases, and amido acids. They are never completely absent from animal and vegetable tissues, but their amount in different substances is very widely variable, some foods being very rich in them, while in others they exist only in traces. Ingested in great excess of the needs of the system, they are likely to cause general disturbance, diarrhoea and albuminuria, while a diet from which they are practically excluded will cause rapid loss of strength, anaemia, great prostration, and greatly diminished resistance to invasion by specific diseases, especially tuberculosis and pneumonia. The animal proteids are more rapidly digested than the vegetable proteids, some of which are largely wasted through imperfect digestion. The proteids, whatever their source, yield in the main the same prod- ucts of digestion, and consequently may replace each other in the diet. The most complete classification of proteids is that of AYroblewski,^ by whom they are divided into three groups as follows : ^ Berichte der deutschen chemischen Gesellschaft, 1897, 30, pp. 3045, 3052. 20 FOODS. Group I. Albuminous bodies closely related to e^g albumin. 1. Albumins. Soluble in water. Egg albumin, serum albumin, laclalbumin, muscle albumin, vegetable albumin. 2. Globulins. Insoluble in water, soluble in dilute salt solution. Egg globulin (vitellin), serum globulin, lactoglobulin, fibrinogen, myo- sin, vegetable globulin. 3. Proteids soluble in alcohol. Substances very slightly soluble in water and salt solutions, soluble in dilute alcohol, some in strong alco- hol. Chiefly of vegetable origin. Very rich in carbon. 4. Albuminates. Products of the action of alkalies on albumins. Slightly soluble in water, easily soluble in alkalies. 5. Acid albumins. Products of the action of acids on albumins. Soluble in very dilute acids and alkalies. Syntonin, etc. 6. Coagulated albumins. Proteid substances coagulated by the action of heat or enzymes. Fibrin, para-casein, etc. Group II. Compound proteids composed of molecules which consist of an albumin group (oj) plus another group, usually of a nour proteid nature. 1. Glyco-proteids. a, plus a carbohydrate group. Mucin, etc. 2. Hsemoglobins. a^ plus a coloring-matter group. 3. Nucleo-albumins. «, plus a nuclein group. 4. Caseins. They contain no true nuclein group. AVith rennet they give a characteristic coagulation reaction. They are not coagulated by heat like albumin nor by enz\'me action like fibrin. Milk casein, legumin, etc. 5. Nucleins. a^ plus a nuclein acid group. 6. Amyloids. Group III. Albuminoids. a, Frame-work substances. 1. Keratins. Constituents of horn. Attacked with difficulty by pepsin and tiypsin. They contain much sulphur and yield much ty rosin. 2. Elastins. Constituents of elastic tissue. They contain less sul- phur and yield less tyrosin. 3. Collagens. Constituents of connective tissue, bone, and cartilage. They contain very little sulphur, and yield no aromatic amido-acids. Gelatin, isinglass, chondrin, collagen, etc. 6. Albumoses and Peptones. Products of hydrolytic splitting of various proteid substances. Their molecules are much smaller tliau those of the albumins. c. Enzymes. Bodies which when present in very small amounts have the property of breaking up very large amounts of certain other sub- stances, including proteids, fats, starches, etc. The albumins are not precii)itated l)y alkaline carbonates, common salt, or dilute acids, but they are coagulated by being lieated to (io"- 73° C. Casein, legumin, conglutin, syntonin, and albuminates, on the contrary, arc not coagulablc by heat, but are precipitated by common salt, sodium acetate, and trisodium phosphate. The albumoses are COMPOSITION OF FOODS. 21 widely distributed throughout the vegetable kingdom, and are found laro-ely in the cereals. In the animal kingdom, they are intermediate products of the action of pepsin on ordinary proteids, becoming even- tually converted to peptones. The coUagens are very rich in nitrogen, but have an inferior nutritive value. Gelatin, for instance, contains 17-18 per cent., while the albumins contain but 16. Fats. — The fats are compounds of the triatomic alcohol, glycerin, with fatty acids, mainly stearic, palmitic, and oleic. These several compounds are known as stearin, pahnitin, and olein. The two first mentioned are solids at usual temperatures, while olein is a liquid. Most fats are combinations of two or all of these substances, and some, as, for example, butter, contain additional glycerides in small amount. Stearin and palmitin being solids, and olein liquid, the consistency of a fat is dependent upon the proportions in which these substances are present. Stearin is a component of most animal fats, but never is found in vegetable fats. The chief constituent of animal fats is pal- mitin, and this occurs also in nearly all vegetable fats. Olein exists in both. Butyrin, caprin, caproin, and caprylin are glycerides of volatile fatty acids present m the fat of milk. Fats consist of carbon, hydrogen, and oxygen, and contain no nitro- gen. The hydrogen and oxygen are not present in the proportions in which they exist in water and in carbohydrates, the oxygen being de- ficient. As taken in food, fats are chiefly in the form of neutral substances, l)ut more or less free fatty acid is always present, and in some foods which have been kept for a time, particularly in well-ripened cheese, fatty acids may be present in a free state in quite large proportion. The fats play an important part in the maintenance of animal heat and mechanical energy. AYhen hard labor is being performed, an excess of fat is instinctively taken. Carbohydrates. — The carbohydrates include the starches, sugars, and cellulose. The Starches, though presenting very different appearances under the microscope according to source, are of equal value as foods, and have the same composition. Starch is insoluble in water, but, heated with it to 72° C, the cells swell and burst, and produce a sort of mucilage. Heated with dilute mineral acids, it is converted into dex- trose. Subjected to the action of diastase, it is converted into maltose. Starch is found almost exclusively in vegetable cells. It forms the chief part of the seeds of the cereals and of the dried residue of certain other vegetable products, such as potatoes. A form of starch known as animal starch or glycogen is found in the liver and muscles, and also in some of the moUusca. Dextrin is an artificial product formed from starch by the action of ferments or of dilute acids and heat. The Sugars are of vegetable and animal origin, and include the foUowino- : 1. Sucrose, cane sugar. A disaccharid. From sugar cane, sorghum, sugar maple, sugar beet, and some other vegetable sources. Insoluble 22 FOODS. in strong alcohol, does not reduce copper ; not directly fermentable. Boiled with dilute acids, is converted to invert sugar, a mixture of dextrose and l8e\'ulose. 2. Dextrose, glucose, grape sugar. A monosaccharid. Found in many fruits and flowers. Formed from cane sugar, maltose, starch, and dextrin by boilmg with dilute acids. In the presence of decom- posing proteids, splits into two molecules of lactic acid. Fermented with yeast, splits into alcohol and carbonic acid. 3. Maltose, malt sugar. A disaccharid. (Two molecules of dex- trose.) Formed from starch by the action of diastase. 4. Lfevulose, fruit sugar. A monosaccharid. Found in honey and various fruits. Rotates the ray of polarized light to the left. Does not form crystals. Isomeric with dextrose. 5. Lactose, milk sugar. A disaccharid. (Dextrose and galactose.) Found only in milk. Behaves like dextrose. 6. Galactose. A monosaccharid. Formed from lactose by boiling with dilute mineral acids. 7. Inosite, muscle sug-ar, phaseomannite. Found in certain animal tissues, as the heart's muscle, and in certain plants, as peas, beans, and grapes. Has no rotatoiy power, does not reduce copper, and is not fer- mentable. It contains the benzene ring, and hence is not a true sugar. In the ]ir€\'^ence of decomposing proteids, it is converted into lactic and butyric acids. Cellulose. — Cellulose, while of value as a food for herbivora, has no nutritive value for man. It is converted to dextrose by boiling with dilute sulphuric acid. Pectin. — Pectin is a substance not unconnnonly classified as a car- be )h\(h-ate. It is composed of carbon, hydrogen, and oxygen, but its precise composition is unknoMU. It is found in various fleshy fruits and in roots, and is believed to be formed from jiectose by the action of vegetable acids. It is known also as vegetable jelly. Pectose. — Pectose is an insoluble substance found in unripe fruits and roots : an earlier stage of pectin. The carbohydrates play an important part in the maintenance of heat and tlie production of force. They lessen the need of fat and form fatty tissue. Excessive ingestion interferes with the metamor- ]>hosis of nitrogenous tissue, causes deposition of fat in excess, and is likely to produce digestive disturbances. Deprivation for a time can be borne, provided the system receives sufficient fatty food, but not otherwise. Organic Acids. — The organic acids exist in foods either in the free state or in comi)ination as salts. In the system they are converted to carbonates, Avhich exercise a most important influence in controlling the alkalinity of the blood and other fluids. Deprivation leads to a peculiar disturbance of the system resulting in scurvy. They include malic, acetic, lactic, oxalic, citric, and tartaric acids. Malic acid is a constituent of apples, pears, and some other fruits. Acetic acid is the essential element of vinegar. Oxalic acid is found in considerable ANIMAL FOODS: MEATS, FISH, EGGS, AND MEAT PRODUCTS. 23 amounts in spinach, tomatoes, strawberries, sorrel, and rhubarb. Lactic acid is present in fresh meats and in milk. The two most important acids are citric and tartaric. The former is found in oranges, lemons, limes, and other fruits ; the latter largely in grapes. Not all vegetables contain these acids, and, therefore, not all have antiscorbutic properties. Potatoes, cabbage, and roots are very effici- ent in this respect, while peas and beans are notable examples to the contrary. Inorganic Salts. — The important inorganic salts taken into the system with food include sodium and potassium chlorides, sodium, potassium, magnesium and calcium phosphates, and compounds of iron. The sulphates are of minor importance and are ingested in only small amounts. The sulphur essential to growth is taken into the system in combination in the proteids. The chlorides keep the globulins of the blood and other fluids of the body in solution, and are the source of the hydrochloric acid of the gastric juice. The phosphates are very essential to the gro^vth of bone and to the nervous system, and iron is needed for the haemoglobin of the blood. Deficiency of calcium and magnesiiun salts leads to rickets and other abnormal conditions. Section 2. ANIMAL FOODS: MEATS, FISH, EGGS, AND MEAT PRODUCTS. The foods of animal origin used by man iuclude the flesh and various organs of the herbivora and swine, domestic and wild fowl, eggs, fish and shellfish, milk and milk products. The flesh of all carnivorous animals except fish is unpalatable, and, therefore, undesirable as food, though under stress of circumstances it may be borne. Thus, during the siege of Paris, about 5,000 cats and 1,200 dogs are said to have been eaten when the food supply had become so meagre that anythiug in the form of flesh was acceptable. In Germany, according to a com- munication of Consul-General Guenther to the State Department at Washington, under date of May 26, 1900, the statistical year-book shows that, on account of the high price of other meats, not only horses, bat also dogs are much used as food. At Breslau, Chemnitz, Dresden, Leipzig, Zwickau, and other places, dogs are slaughtered extensively for this purpose and regularly inspected. PirP reports that in Saxony during 1894, 295 ; in 1895, 388 ; in 1896, 399 ; and in 1897, 474 dogs were slaughtered and inspected. In Dessau, between 1893 and 1898, the number averaged 251 yearly, and inspection showed that one in 202 was trichinous. According to Tempel,' of 289 killed at Chemnitz during 1897, 1.391 per cent., and of 147 killed during the first half of the year 1898, 2.04 per cent, were found to be trichinous. The meat is eaten chiefly in the roasted state, but also, in many parts of Saxony, raw, but highly sea- ^ Zeitschrift fiir Fleisch- und Milchhygiene, X., No. 1. 2 Ibidem, IX., Xo. 1. 24 FOODS. soned. The same animals are commonly eaten by the Chmese, and the Canada lynx and the skunk are rated as delicacies by the North Ameri- can Indians. MEATS. The value of meat as food depends upon the presence of proteids, fat, and mineral salts. The nitrogenous extractive matters (creatin, etc.), sometimes called '' meat bases," formed by cleavage of the pro- teids, give flavor, but have little value as foods. The carbohydrates play but an insignificant part, being present chiefly as muscle sugar and to only a very small extent. All meat, however lean, contains fat, some of which is visible and some indistinguishable from the muscle fibres by which it is surrounded. The visible fat varies widely in amount. Very fat beef may contain considerably more than a quarter of its weight of visible iat, and fat pork meat more than a half, while chicken and veal contain comparatively little. The content of water varies very widely and in general may be said to be governed by the richness in fat, for, while the proteids are fairly constant in amovmt, the remainder is almost wholly water and fat, and the greater the amount of the one, the less the amount of the other. The less fat a meat contains, the less, therefore, its relative nutritive value. Digestibility. — \\'hile the amount of nutriment contained in meats chiefly determines their food value, the latter is to no inconsiderable extent dependent upon the ability of the alimentary tract to digest and absorb them. Gastric digestion is by no means to be accepted as a measure of the true digestibility of a food, and the same is true of the results of artificial laboratory experiments ; hence many of the accepted statements bearing on this subject, based upon the oft-quoted experi- ments on Alexis St. Martin and upon test-tube observations, may be wholly disregarded. Raw meat is digested more easily, but less completely, than that which has undergone the process of cooking, and rctasted meat is more completely digested than that which has been boiled. Fat meats, as beef and mutton, and especially pork, require more time for digestion than those which, like chicken and veal, contain but little fat. In general, it may be said that meats are assimilated more easily than vege- table foods. Flavor. — The flavor of meats depends largely upon the nature and amounts of nitrogenous extractives which they contain, and is greatly modified by the condition of the animal when killed, its age, sex, and tlie character of its food. The high flavor of birds and game is due to the richness in extractives, while in the case of meats deficient in these substances, as, for example, mutton and pork, the flavor is due largely to their contained fats. ]Most meats are im])roved in flavor by being kept for a time, during which, additional flavors, due to decompo- sition products similar to the extractives, are (levelo]ied. The meat of young animals is flavored less highly than that of adult's, and MEATS. 25 that of females tlian that of males, though in the case of the goose this condition is reversed, and m that of swine no diiference is ob- servable. Asexualization by spaying or castration produces a fatter, more tender, and better flavored meat. Thus, the flesh of oxen is far pre- ferable to that of bulls or cows, and that of capons and poulards to that of cocks and hens. Texture. — Whether a given meat is tough or tender depends upon the character of the walls of the muscle tubes and upon the amount of connective tissue present. The tube walls are thin and delicate, and the connective tissue is small in amount in the young and well-fed, but as the animal becomes older or is made to work, the tubes thicken and become hard, the connective tissue increases in amount, the fat may diminish, and the result is a coarser flesh. Very young animals have a very watery, gelatinous, and flavorless flesh. The textiure of meat undergoes very considerable change after slaughter. When freshly slaughtered, it is tender and juicy, but as rigor mortis supervenes, it becomes hardened and tough. The stage of rigor is succeeded by the first stage of decomposition, during which lactic acid is formed. This acts upon the connective tissue and causes softening and tenderness, and as the process of decomposition proceeds within proper limits, increase of flavor is developed. Effects of Cooking'. — When meat is cooked, the connective tissue is softened, the bundles of fibrillse are loosened from each other, the albiunin is coagulated, flavors are improved and new ones developed, parasites and micro-organisms are destroyed, and the whole mass is rendered more acceptable to the eye and palate. In the process of roasting or broiling, considerable shrinkage due to loss of water occurs. The heat to which the meat is subjected should be sufficiently intense to produce speedy coagulation of the exterior and prevent the meat juices from becoming di'ied up. In order that the surface shall not be burned, the meat must be basted from time to time with hot melted fat, which forms a protective coating. The heat employed should be less intense with large joints than with small ones, since before the heat can pene- trate well into the interior, the outer parts will become burned. In boiling, the temperature of the water into which the meat is im- mersed varies according to the object sought. If a rich broth is de- sired, the meat is placed in cold water, which then is heated gradually. During the heating process, the soluble albumins together with a por- tion of the salts and the extractives are dissolved out. When the tem- perature reaches 134° F., the albumin begins to coagulate, and above 160°, the connective tissue is changed to gelatin and dissolved. The solution of certain of the constituents is assisted by the small amounts of lactic acid formed. If, on the other hand, it is desired to have the juices and flavors re- tained within the mass, the meat should be plunged into boiling water, which quickly coagulates the albumins at the surface and causes thereby the formation of a protective coating. After this is formed, the tem- 26 FOODS. perature should be lowered to about 180° F,, for otherwise the meat be- comes tough, even to the center. The shrinkage in meat that has been j)roperly boiled amounts to from 20 to 40 per cent, of its weight. In frying, the meat is dropped into veiy hot fat, as lard or vegetable oil, Avliich causes speedy coagulation of the surface, such as is brought about in the process of boiling, whereby all the flavors and juices are retauied. It is essential that the fat be very hot, since otherwise it will penetrate the tissues and cause the meat to become greasy and un- palatable. In stewing, the meat is cut into small pieces and placed in cold water, which then is heated slowly to about 180° F.; at which temperature the whole is kept for several hours. If heated above 180°, the meat becomes tough, stringy, unpalatable, and of diminished digestibility. Characteristics of Good Meat. — ISIeat should have a uniform color, ueitlier abnormally pale nor inclined to purplish. It should have little or no odor, and such as it has should give no disagreeable impression such as the sickly cadaveric smell characteristic of diseased or decom- posing flesh. It should be firm and elastic, and should not pit nor crackle on jiressure. On being handled, it should scarcely moisten the fingers, and Avith keepmg, the exterior should become dry rather than wet. There should be no evidence whatever of the presence of parasites. Beef has a bright red color and a marbled appearance, due to the presence of fat between the bundles of muscular fil)ers. This marbling is much less a]iparent in the flesh of animals that have not been well fed and of old cows and bulls. Bull meat is darker than that of oxen and cows, and is coarse, stringy, and of strong flavor. Veal is much paler than beef and less firm to the touch, and coming from a verv young animal, '' bob-veal," it is flabby and watery, and its fat has a tallowy a])]iearauce. Mutton should be of a dull-red color and firm to the touch. Its fat is white, sometimes yellowish, and hard. Lamb is somewhat less firm to the touch and has a decidedly lighter color than mutton. Pork is much less firm to the touch than beef and mutton, and its fat is quite soft in comparison. Horse meat, the use of Avhieh is increasing rapidly abroad and U\ a much greater extent in this country than is conunonly believed, is darker and coarser than beef and jiossesses a very dift'erent odor. The fat is yellow and oily and has a rather disagreeable odor. The flesh of birds is not marbled like that of mammals. That of wild fowl that feed on fish has a strong flavor, which is not improved by keeping. Comparative Digestibility of Meats. — Beef is commonly and cor- rectly regarded as one of the most digestible of meats, but according to the experience and testimony of mtiny victims of dyspepsia it is inferior in this respect to mutton. Pork is, without doubt, digested with greater difficulty than any other meat, on account of its high content of fat. The evidence as to veal is most conflicting, some holding that MEATS. 27 it is digested very easily, while others maintain the contrary view. Certain it is that many persons bear it vers' badly. The white meat of chickens, fowls, and turkeys, is more delicate and is digested more easily than the dark meat, probably by reason of its smaller amount of fat. The flesh of ducks and geese is harder, richer, and more diffi- cult of digestion. Game birds are less fat than poultry and are often much better borne. Their habits of life are unfavorable to the depo- sition of much fat. Liver, kidneys, and heart are generally regarded as unsuitable as foods for persons with weak stomachs, but tripe and sweetbreads are usually easily borne. " Red Meat " and " White Meat." — The prohibition of red meats (beef, mutton, venison) to patients with gouty and rheumatic tendencies dates from the time of Sydenham, whose chetetic rules allowed only the white meats (veal, goat, yoimg pig, chicken) and fish to such persons. Today, many practitioners extend this prohibition to those with diseases of the stomach, intestines, and kidneys, and various neuroses. The foundation of this prejudice against the red meats is the supposed presence in them of a greater percentage of the nitrogenous extractives (creatin, xanthin, guaniu, etc.), which are believed to exert injurious action in two ways : First, locally, by irritating the kidneys during the process of their elimination from the body; and second, in cases of impaired fiuictional activity of the kidneys, by causing systemic intoxication. Unfortunately, however, for the stabihty of this belief, exact analysis has shoT^m that the very small amounts of these sub- stances present are practically the same in both red and white meats, with the single exception of venison, which contains them not, as would be surmised, in highest percentage, but, in fact, in lowest. Further- more, these extractives are not eliminated as such, but as the normal ulti- mate product of metamorphosis, urea. It has been supposed, too, that the non-nitrogenous extractives (lactic, but^Tic, and acetic acids, etc.) are present to a greater extent in red than in white meats and may cause disturbance ; but as a matter of fact, these are present in extremely small amounts in both red and white meat, and cannot possibly be regarded as harmful, in view of the fact that appreciable amounts exert no influence on the svstem. Composition of Meats. — In the following tables, showing the com- position of the edible portions of meats, the figures given are taken, unless otherwise stated, from Bulletin Xo. 28 (re^'ised edition) of the Office of Experiment Stations of the U. S. Department of Agriculture : The Chemical Composition of American Food Materials.^ ^ GoTemment Printixig Office, "Washington, 1899. 28 FOODS. BEEF. Cut. Brisket, medium fat . . . Chucii, including shoulder. very lean lean medium fat fat very fat Average Chuck rib, very lean . . . lean medium fat fat Average Flank, very lean .... lean medium fat . . . fat very fat Average Loin, very lean lean medium fat fat very fat Average Porterhouse steak .... Sirloin steak Top of sirloin Tenderloin Ribs, very lean lean medium fat fat very fat Average Rib rolls, very lean . . . lean medium fat fat Average Round, very lean .... lean medium fat fat very fat Average Round, second cut .... Rump, very lean .... lean medium fat fat very fat Average 0! !2 'S o .a 00 < 3 54.6 15.8 28.5 0.9 1495 1 73.8 22.3 3.9 0.8 580 9 71.3 20.2 8.2 1.0 720 4 68.3 19.6 11.9 0.9 865 4 62.3 18.5 18.8 0.9 1135 2 53.2 17.2 29.0 0.9 1555 13 65.0 19.2 15.4 0.9 1005 1 75.8 22.2 1.4 1.1 470 11 71.3 19.5 8.3 1.0 715 7 62.7 18.5 18.0 1.0 1105 2 52.0 16.6 31.1 0.8 1620 21 66.8 19.0 13.4 1.0 920 3 70.7 25.9 3.3 1.2 620 3 67.8 20.8 11.3 1.0 865 5 60.2 18.9 21.0 0.9 1240 3 54.2 17.1 28 4 0.8 1515 2 34.7 14.0 51.8 0.7 2445 16 59.3 19.6 21.1 0.9 1255 3 70.8 24.6 3.7 1.3 615 12 67.0 19.7 12.7 1.0 900 32 60.6 18.5 20.2 1.0 1190 6 54.7 17.5 27.6 0.9 1490 3 49.7 17.8 32.3 0.9 1695 56 61.3 19.0 19.1 1.0 1155 7 60.0 21.9 20.4 1.0 1270 21 61.9 18.9 18.5 1.0 1130 1 42.2 13.8 43.7 0.8 2100 6 59.2 16.2 24.4 0.8 1330 4 65.7 21.9 1.1 0.7 455 6 67.9 19.6 12.0 1.0 870 15 55.5 17.5 26.6 0.9 1450 9 48.5 15.0 35.6 0.7 17S0 1 45.9 14.6 38.7 0.6 1905 35 2 57.0 17.8 24.6 0.9 1370 73.7 20.8 5.0 1.0 600 3 69.0 20.2 10.5 1.0 820 4 63.9 19.3 16'.7 0.9 1065 2 51.5 17.2 31.3 0.8 1640 11 64.8 19.4 15.5 0.9 1015 6 73.6 22.6 2.8 1.3 540 31 70.0 21.3 7.9 1.1 730 18 65.5 20.3 13.6 1.1 950 5 60.4 19.5 19.5 1.0 1185 2 55.9 18.2 26.2 0.8 1445 62 67.8 20.9 10.6 1.1 835 2 69.8 20.4 8.6 1.1 740 4 71.2 23.0 5.1 1.2 645 4 65.7 20.9 13.7 1.0 965 10 56.7 17.4 25.5 0.9 1400 5 47.1 16.8 35.7 0.8 1820 1 40.2 15.0 44.3 0.8 2150 24 57.9 18.7 23.1 0.9 1325 MEATS. 29 BEEF.— Continued. Cut, Beef heart Kidney (carbohydrates 0.4) Liver (carbohydrates 1.5) . Marrow Tongue Lungs Suet Koast beef (cut not specified) Sirloin steak, baked .... Broiled tenderloin .... Bound steak Canned corned beef .... Canned roast beef Canned whole tongue . . - Canned tripe Corned beef (all cuts) . . . Tongues, pickled Tripe (carbohydrates 0.2) O w "S 2 'S g ■s f^ < 111 a ■'- <^ 2 62.6 16.0 20.4 1.0 1160 3 76.7 16.6 4.8 1.2 520 6 71.2 20.7 4.5 1.6 605 1 3.3 2.2 92.8 1.3 3955 3 70.8 18.9 9.2 1.0 740 1 79.7 16.4 3.2 1.0 440 9 13.7 4.7 81.8 0.3 3540 7 48.2 22.3 28.6 L3 1620 1 63.7 23.9 10.2 1.4 875 6 54.8 23.5 20.4 1.2 1300 18 63.0 27.6 7.7 1.8 840 15 51.8 26.3 18.7 4.0 1280 4 58.9 25.9 14.8 1.3 1105 5 51.3 19.5 23.2 4.0 1340 2 74.6 16.8 8.5 0.5 670 10 53.6 15.6 26.2 4.9 1395 2 62.3 12.8 20.5 4.7 1105 4 86.5 11.7 1.2 0.3 270 POEK. Cut. Ham, fresh, lean medium fat fat visible fat largely removed . Loin (chops), lean medium fat fat Tenderloin f^houlder Feet Head cheese Kidney Liver (carbohydrates 1.4) . . . Ham, smoked, lean medium fat fat smoked, boiled Shoulder, smoked, medium fat . fat Salt pork, fat lean ends ........ Bacon, smoked, lean medium fat S S3 2 10 5 3 1 19 4 11 19 8 3 2 1 3 14 4 2 o 2 7 4 2 17 60.0 53.9 38.7 64.5 60.3 52.0 41.8 66.5 51.2 50.7 43.3 77.8 71.4 53.5 40.3 27.9 51.3 45.0 26.5 7.9 19.9 31.8 18.8 25.0 15.3 12.4 19.2 20.3 16.6 14.5 18.9 13.3 8.3 19.5 15.5 21.3 14.4 28.9 50.0 16.2 19.0 30.1 44.4 ]3.0 34.2 17.4 33.8 4.8 4.5 19.8 16.3 14.8 20.2 15.9 15.1 1.9 8.4 15.5 9.9 20.8 38.8 52.3 22.4 32.5 53.6 86.2 67.1 42.6 67.4 1.3 0.8 0.7 0.9 1.0 1.0 0.7 1.0 0.8 0.4 3.3 1.2 1.4 5.5 4.8 3.7 6.1 6.7 5.2 3.9 5.7 ILO 4.4 Hi 1075 1505 2345 1040 1180 1580 2145 900 1690 1090 1790 490 615 1245 1940 2485 1320 1665 2545 3670 2985 2085 3030 30 FOODS. VEAL. Cut. Breast, lean . . medium fat Leg, lean . . . medium fat cutlets . . Loin, lean . . medium fat fat ... . Heart .... Kidney .... Liver .... 10 3 5 6 2 1 72.1 66.0 73.5 70.0 70.7 73.3 69.0 61.6 73.2 75.8 73.0 21.7 19.5 21.3 20.2 20.3 20.4 19.9 18.7 16.8 16.9 19.0 5.6 14.0 4.1 9.0 7.7 5.6 10.8 18.9 9.6 6.4 5.3 1.1 1.0 1.2 1.2 1.1 1.2 1.0 1.0 1.0 1.3 1.3 5 c «= "3 3.2 > 2 !:! 0) IH oj 640 955 570 755 705 615 825 1145 720 585 575 MUTTON. Cut. Hind leg, lean medium fat fat Loin, medium fat without kidney and tallow fat without kidney and tallow .... very fat without kidney and tallow . . Fore quarter Hind quarter Koast leg, cooked Kidney Liver (carbohydrates 5.0 J O xa u "^ u %h ^ 3 67.4 11 62.8 1 55.0 13 50.2 3 43.3 1 30.8 10 52.9 10 54.8 2 50.9 1 69.5 2 61.2 LAMB. Cut. Hind leg, medium fat .... fat very fat Loin, without kidney and tallow Fore quarter Hind {[uarter Broiled chops Roast leg % ■o 1 rk, especially those which have undergone a process of cooking before being stuffed, are not always beyond suspicion, for the trichinous jwrk condemnearation of sausage : it is inexjiensive, and it lends a firmness which increases as the sausage is boiled. A number of chemical processes have been pro- MEATS. 33 posed for its detection, and among them the following : Humbert ^ boils about 50 grams for an hour in 200 cc. of water, adds about 10 CO. of strong nitric acid after cooling, and then filters through paper. Some of the filtrate, contained in a test-tube, is then treated with a volume of freshly prepared iodine water by careful addition, so that the liquids do not mix. At the line of junction, a reddish violet zone will be seen, if the specimen contained any considerable proportion of horse meat. This reaction of iodine upon glycogen, which is a normal constituent of horse flesh, forms the basis of a number of processes for the detec- tion of this meat, but it is criticised by Mebel ^ and others as uncertain, since glycogen is present also in liver, meat extracts, and veiy young veal, and since dextrin resulting from added starches gives a similar coloration. Another process suggested differs from the above in that acetic acid is added durmg the boilmg process, and the reagent em- ployed is a stronger preparation of iodine, being a 10-per-cent. solution with potassium iodide. In the manufacture of all grades of sausage, scrupulous care should be observed to secure cleanliness of the casings, which require more thorough treatment than the mere passage of water through them. Dr. Schilling ^ reports the examination of prepared intestines which yielded 5 grams of excrement per meter. O^wdng to the occurrence of a gray color, which is said by Meyer ^ to be due to the passage of salt by diffusion from the contents to the casing, which is rich in water and poor in salt, the commercial value of certain varieties of sausage is impaired, and hence it becomes neces- sary to apply artificial colors, or so to treat the stuffing that the change in color shall not occur. The very red appearance which so often sug- gests the presence of coal-tar products may be due to the action of harmless preservatives, like niter, or the haemoglobin of swine blood. In such case, the extract with alcohol, glycerin, or amyl alcohol will not dye wool, and the color cannot be precipitated as a lake. According to J. Haldane,^ the red color of cooked salted meats is due to the presence of XO-hsemochromogen produced by the decomposi- tion of jSfO-hsemoglobin, which is formed by the action of a nitrite on the NO-hjemochromogen in the absence of oxygen and presence of re- ducing agents. The nitrite is formed by reduction, within the raw meat, of the niter used in salting. Certain of the artificial sausage colors, as " Orange II.," the sodium salt of ^9-naphtholazobenzene, are extracted easily with acidulated water, and will dye woolen fibers immersed therein. ^ Kecueil de Medicine Veterinaire, 1895. ^ Zeitschrift fiir Fleish- nnd Milch hygiene, 1895, p. 86. ^ Deutsche medicinische Wochenschxift, 1900, p. 602. * Chemiker Zeitung, 1900, p. 3. ^ Journal of Hygiene, 1901, Vol. I., p. 115. 34 FOODS. FISH. In the ordinary sense of the word, fish iuchides, in addition to the varieties of fish in its narrow sense, molhisks (chims, oysters, mussels, etc.) and crustaceans (lobsters, crabs, crawfish, and shrimps). Many prejudices have existed from the earliest times, and to a cer- tain extent still exist, against the use of fish in the diet. The ancient belief that a fish diet is a common cause of leprosy still obtains to a certain extent, even among enlightened people, in spite of all scientific evidence to the contrary. Thus, ]\Ir. Jonathan Hutchinson ' main- tains that this disease is so caused, especially if the fish is poorly cooked or partially decomposed. He asserts that the disease prevails near the sea and is disappearing before the advance of agriculture ; but op- posed to this i» the fact that, in the interior of India, the disease is very common among people whose religion forbids the use of fish, and who cannot obtain it even if it were desired. Some varieties of fish cannot be eaten, because of their inherent poisonous properties. But few of these are, however, found north of the tropics. Some of them are always poisonous, and others only at certain times ; and in some cases, individuals of certain species may be and others may not be noxious. Some varieties of fish are the hosts of parasites, some of which are injurious to man, but unless eaten in the raw state they are not likely to produce harm. There is a belief that fish is particularly valuable as a brain food, on account of the sujiposed high percentage of jihosphorus that it contiiins. The amount of ]ihos})horus is, however, so far as is kno'\\'n, no higher in fish than in meat and, moreover, this element is no more essential to the brain and nervous system than any others which are present. If therQ were any tnith in this conmion belief, we should expect naturallv to find men of commanding intellect among those whose diet consists mainly of fish, but, as is well known, such people are of a low order of intelligence, though not by reason of their diet. In spite of the large amount of nutriment contained, fish has not the same satisfving properties that belong to meats, but it is an exceed- ingh- valuable food, and in many parts of northern countries is the principal animal food. The flavor of fish is influenced greatly liy the nature of the food supply and by the content of fat. Generally speak- ing, salt-water fish from deep water, where the current is strong, are better than those from shallow, warm, and quiet water, and both salt- Avater and fresh-water fish taken from rocky and sandy bottoms are preferred to those from muddy bottoms. Condition is dependent upon a variety of circumstances. Some fish are regarded most highly during their spawning season (shad and smelts), while others are looked u])on with disfavor at this period. Fish caught by the gills in gill nets die slowly, but decompose rapidly, and such are of inferior flavor and value. Fish taken from the water ' Archives of Surgery, April, 1899. FISH. 35 alive and killed at once remain firm and retain their flavor longer than those allowed to die slowly. Digestibility. — So far as is known^ the digestibility of fish and meat is about equal, but, as is true of meats, different varieties of fish are differently digestible. Thus, those which contain the greatest per- centages of fat, as salmon, eels, and mackerel, are the most difficult of digestion. The processes of drying, smoking, salting, and pickling lessen digestibility, and fish so treated are, in consequence, less suited to the needs of invalids and dyspeptics. The moUusks are regarded generally as most digestible, while the crustaceans are believed to tax the digestive powers much more than any other animal food. Many persons are unable to digest them in any form, and others who suffer no inconvenience, so far as digestion is concerned, are obliged to prac- tise self-denial, because of idiosyncrasy, which manifests itself in dis- tressing eruptive disorders of the skin, dizziness, and other nervous symptoms. Lobster and crabs are much alike, but the former is less likely to disagree. The claws of the lobster are more tender and deli- cate than the tail, which is firmer and much closer grained. Shrimps are rated generally as more difficult of digestion than lobsters and crabs, but with many they are borne more easily. The moUusks are more digestible ui the raw state than when cooked. The comparatively tough part of the oyster, the adductor muscle, is very trying to some persons, and for such it is best removed and rejected. Keeping" Qualities. — Fish flesh differs very greatly from meats in keeping qualities. While the latter are improved up to a certain point by hanging, fish should be eaten while fairly fresh, since decomposition sets in very quickly. Some varieties, as halibut, cod, haddock, and turbot, may be kept a week or more when properly cared for, while others begin to deteriorate almost immediately. So long as the flesh is firm and stiff*, all fish is edible, but when it is crushed readily by gentle pressure between the fingers, it is unsound and should be re- jected. MoUusks and crustaceans decompose very quickly, and the latter when boiled a short time after natural death have but little flavor. Composition. — In proteids, fish rank nearly as high as meats, but they are very much poorer in fat, only a few varieties yieldmg over 10 per cent. These include salmon, turbot, lamprey eels, eels, butterfish, lake trout, and herring, and are followed by shad and Spanish mack- erel, with over 9 per cent. The great majority of species contain less than 5 per cent., and many of the commoner kinds even less than 1 per cent. In fact, most fish flesh yields more mineral matter than fat. Shellfish are fairly rich in proteids and contain notable amounts of carbohydrates, but they are very poor in fat. The chemical composition of the edible portions of many varieties of American fish is given in the following table compiled from the report of the U. S. Commissioner of Fish and Fisheries for 1888 : 36 FOODS. Kinds of fish. California salmon Eel Salmon Spanish mackerel Lake trout Whitefisli Butterfish Shad Lamprey eel Turbot ■. Mackerel Herring Pom])ano Alewife Small-mouthed black bass . Halibut Sheepshead \\'hite perch Pollock Cisco Muskellunge Striped bass I>rook trout Eed snapper Blueiish Large-moutlied black bass . Small-moutlied red-horse . . Sturgeon Skate Weakfish Bhu-kfisli Smelt Kingfish Sea bass (Jrouper Yellow perch Pike perch, Wall-eyed pike Pickerel Pike perch, Gray pike . . . Haddock . . . ". Tomcod Red bass Cnsk Cod Hake Common flounder Winter flounder Spent fish. Spent salmon Spent land-locked salmon . Proteids. per ct. 17.46 15.82 20.77 20.97 18.22 22.06 17.81 18.55 14.93 12.92 18.77 18.19 18.65 19.17 21.50 18.35 19.54 19.03 21.65 19.12 19.63 18.31 18.97 19.20 19.02 19.24 17.90 17.96 18.82 17.45 18.47 17.36 18.66 18.75 18.80 18.49 18.42 18.64 17.26 17.10 17.08 16.68 16.92 16.00 15.24 13.82 14.01 18.52 17.24 Fat. per ct. 17.87 18.74 12.09 9.43 11.38 6.49 11.03 9.48 13.29 14.41 8.21 8.02 7.57 4.92 2.44 5.18 3.69 4.07 0.78 3.48 2.54 2.83 2.10 1.03 1.25 0.96 2.35 1.90 0.93 2.39 1.35 1.80 0.95 0.49 0.60 0.70 0.47 0.50 0.76 0.26 0.38 0.53 0.17 0.30 0.67 0.69 0.44 3.60 2.98 In flesh Ash. per ct. 1.06 0.93 1.38 1.50 1.26 1.62 1.14 1.35 0.66 1.28 1.40 1.69 1.00 1.47 1.24 1.05 1.22 L19 1.55 1.25 1.57 1.16 1.21 1.31 1.27 1.19 1.19 1.43 1.43 1.19 1.08 1.68 1.18 1.44 1.15 1.29 1.37 1.18 1.13 • 1.25 0.99 1.23 0.90 1.24 0.98 1.28 1.20 1.14 1.24 Total. I Water, i^^^l value per pound. per ct. 36.39 35.49 34.24 31.90 30.86 30.17 29.98 29.38 28.88 28.61 28.38 27.90 27.22 25.56 25.18 24.58 24.45 24.29 23.98 23.85 23.74 22.30 22.28 21.54 21.54 21.39 21.44 21.29 21.18 21.03 20.90 20.84 20.79 20.68 20.55 20.48 20.26 20.32 19.15 18.61 18.45 18.44 17.99 17.54 16.89 15.79 15.65 23.26 21.46 per ct. 63.61 64.51 65.76 68.19 69.14 69.83 70.02 70.62 71.12 71.39 71.62 72.10 72.78 74.44 74.82 75.42 75.55 75.71 76.02 76.15 76.26 77.70 77.72 78.46 78.46 78.61 78.56 78.71 78.82 78.97 79.10 79.16 79.21 79.32 79.45 79.62 ' 79.74 I 79.68 80.85 I 81.39 81.55 I 81.56 ! 82.01 I 82.46 83.11 1 84.21 84.35 76.74 78.54 Calories. 1,080 1,085 895 790 820 685 795 745 840 850 695 675 665 565 505 560 520 525 435 505 470 460 440 400 405 400 430 415 390 425 400 400 385 370 375 375 360 370 355 330 335 335 320 310 310 285 280 495 445 MEAT AND FISH AND PARASITIC DISEASE. 37 Fish. Edible portion. H Desiccated codfish (fish meal) Salt codfish Salt mackerel Smoked haddock Smoked halibut Smoked herring Canned salmon Canned mackerel Canned tunny per ct. 74.46 2L42 18.88 33.68 20.57 36.44 20.06 19.91 21.52 per ct. 1.90 0.34 25.12 0.17 15.03 15.82 15.70 8.68 4.05 per ct. 5.41 1.62 2.59 1.53 2.06 1.53 1.32 1.30 1.69 per ct. 84.75 46.42 56.99 27.44 50.62 65.45 38.12 31.82 27.26 per ct. 15.25 53.58 43.01 72.56 49.38 34.55 61.88 68.18 72.74 Calor. 1,465 410 1,410 445 1,015 1,345 1,035 735 570 per ct. 2.88 23.04 10.40 2.06 12.96 11.66 1.04 1.93 Fish. '3 'S A "3 o S < ^ Ph A 9.78 2.05 5.89 1.98 19.70 1.48 0.03 0.75 2.13 4.39 7.41 2.07 3.95 1.31 14.74 14.55 1.79 2.94 2.76 22.04 17.73 2.89 1.59 3.16 25.37 11.59 0.74 7.21 2.22 21.76 16.70 1.27 4.14 2.33 24.44 12.51 1.67 5.42 1.73 21.33 14.75 0.17 3.38 1.38 19.68 14.49 1.84 1.71 18.13 1.07 2.47 16.64 1.96 3.13 15.80 1.55 1.94 25.38 1.00 2.58 21.23 3.47 1.02 19.84 0.53 1.20 Oysters Oyster liquor .... Canned oysters . . . Long clams Long clams (canned) Bound clams .... Round clams (canned) Mussels Scallops Lobster Lobster (canned) . . Crab Crab (canned) . . . Shrimps (canned) . . Tei-rapin Green turtle .... 80.30 95.61 85.26 77.96 74.63 78.24 75.56 78.67 80.32 Meat and Fish and Parasitic Disease. Man is often the host of parasites through ingestion of infested meat and fish. Of these, the ' most common is the tapeworm, of which at least ten species are known, though only three have been demonstrated as having any connection with food. These are Taenia saginata (T. mediocanellata) due to measly beef, Tsenia solium to measly pork, and Bothriocephalus latus to infested sturgeon, pike, perch, and salmon. The latter is very rare in this country, though not uncommon along the Baltic. Of the large number of worms which infest fish, this is the only one known to be conveyed to man. It is killed very quickly if the fish is cooked properly. The larval forms of T. saginata and T. solium exist in beef and pork respectively as Cysticercus bovis and C cellulosae. The latter is found rarely also in mutton. Meats infested with these forms are 38 FOODS. known as " measly " or '^ measled," and the animals from which the meats are derived are known as the intermediate hosts, man being the final host. The life history of the tapeworm is much more simple than that of many other parasites. Beginning with the adult tape- worm in man, the cycle is as follows : From each individual segment, which is possessed of a complete reproductive system, great numbers of eggs are discharged. The latter are expelled from the intestine with the feces, and some of them eventually may enter the digestive tract of cattle or swine through the food or water. In the intestine, the embryos become liberated from the eggs, and they then make their way in large numbers to the muscular tissues, brain, liver, and other parts, where they come to rest and develop as bladder Avorms. The bladders are variable in size, the smallest being about -^ inch, and the largest about 1^ inch across, and, in the flesh, are embedded between the muscular fibers. The living animal shows nothing in its appearance to indicate the presence of the parasite, excepting when the cyst can be seen in the under side of the tongue or between the tongue and the lower jaw. If now the animal is slaughtered and the meat is eaten raw or imperfectly cooked, the Cysticercus is freed from its enveloping capsule and pro- ceeds to develop into an adult tapeworm, and the cycle is complete. The Cysticercus bovis dies within two or three weeks after the slaughter of its host, and, therefore, measled beef, kept for three weeks in cold storage, becomes incapable of 2)roducing harmful effects. It is killed within 24 hours by pickling solutions of common salt, when brousfht into intimate contiict. The Cvsticercus cellulosae lives rather longer in cold storage : probably a month or more, Both of these larvje are killed by exposure to a temperature of 140° F. for o minutes, and as this is lower than the temperature in the interior of well-cooked meat, it is necessary only to make sure that the meat is properly cookcfl to escape danger. Neither parasite is destroyed by ordinary smoking or salting, but both are killed by hot smoking. The sale of measled meat is permitted in many countries of Europe, but it must be sold as such and only in specially designated places. It finds a ready market at a low price, and the purchasers are warned that it should be cooked thoroughly. According to Virchow, since the introduction of systematic meat inspection, the proportion of tapeworm in the human subject dissected in Berlin has fallen from 1 in -U to 1 in 280. A parasite of far greater importance is the Trichina spiralis, Avhich is found almost exclusively in pork and only occasionally in the fle-h of other mammals and of birds and frogs. Trichinae are small, thread- like worms, much longer than one would suppose on pa^sillg examina- tion of fiiu'ly thick microsco])ic preparations, since they are coiled with several turns within the minute cysts in which they are lodged. In Plate I., Fig. 1, is shown one of the parasites in the free state. In the pig the worm infests both the fat and the voluntary muscles, but chieHy the latter, and esjKH-ially the diaphragm, the intercostals, and the nuis- cles of the jaw. When encapsulated in the flesh, their location may be PLATE I. Fig. 1. Free Trichina. X 38. Fia. 2. Trieliinse in Human Muscle. X 76. MEAT AND FISH AND PARASITIC DISEASE. 39 evident to the naked eye as small white specks. Thin sections of muscles, treated a few minutes in weak caustic potash solution (1 : 10), are soon made sufficiently clear to reveal the coiled worm under a lens of low power. If very much fat is present or if the capsule has become calcareous and thick, it may be necessary to employ ether or acetic acid before applying the potash. When the parasites are very numerous, the flesh is both speckled and pale. Our first knowledge of the serious effects which may result from eating infested pork dates back only to 1860, although the parasite had been discovered in the muscles of a human subject by James Paget in 1833 and named by Richard Owen. It long was regarded as a harm- less parasite and curiosity, and its effects were mistaken for rheuma- tism, typhoid fever, and other diseases of common occurrence. The case which finally revealed its capacity for harm was that of a young woman admitted to the hospital at Dresden suffering, it was supposed, from typhoid fever. In a short time, a train of symptoms quite unlike those of that disease appeared, the most marked one being very acute pain involving the entire muscular system, and intensified on attempting to move. On account of the agony induced, extension of the arms and legs was quite impossible. Pneumonia supervened, and in a few days the victim died. The autopsy revealed the parasite in vast numbers in the muscles, and this led to further investigation, which showed that, four days before the first symptoms of illness appeared, she had eaten freshly killed pork. Some of this was secured in the form of ham and sausage, and examination demonstrated the presence of the parasite. The first extensive outbreak which caused the disease to be looked upon as one of great importance occurred in Prussia in 1863, when more than 20 persons died within a month after a dinner in which 103 had participated, and at which smoked sausages made from an infested pig had been served. The parasites were discovered in the muscles of those who died and in the sausages that remained. Since that time, it has been customary in most large slaughtering establishments to examine pork for evidence of the parasite, before passing it as fit for food. But examination is not always a safeguard, even in countries where it is observed most carefully. In Germany, for instance, where all meats are supposed to be examined with scrupulous care, particu- larly those from the United States, the disease is very common. In 1883, on account of the alleged dangers which lurked in Ameri- can meats, importation was interdicted for a time, but in the succeed- ing 15 years there were in Prussia alone 3,003 cases and 207 deaths, not one of which could be traced to American meat either salted, pickled, canned, or made into smoked sausages. Over 40 per cent, of the cases were traced to European meat which had been passed as free from trichinse, and the rest to European meat which had been found to contain the parasite, but had, nevertheless, been handled by the trade. During 1899, the parasites were found by the microscopists of the U. S. Department of Agriculture in 41,659 of the 2,227,740 hogs examined. 40 FOODS. It is probable, as stated by Charles W. Stiles/ who collected about 900 reported cases which had occurred in this couutiy during the 36 years 1860-1895, that the disease is more common in the United States than generally is supposed. It is accepted commonly that 1 or 2 per cent, of dissecting-room subjects show evidence of the parasite, but it would appear from the investigations of H. U. Williams ^ that this estimate is much too low, for careful examination of 505 cadavers taken at random showed old encapsulation and calcifi- cation in no less than 27 instances, or 5.34 per cent. The birthplaces of the subjects included all of the most important countries of Europe and North America, but the number of cases examined was too small to admit of accurate conclusions as to the influence of nationality upon the frequency of the disease. It is evident that many cases of trichinosis escape detection, and that, as Williams points out, estimates have been based on nakcd-cve diasrnosis. In spite of the danger of eating trichinous meat, there are those who are not deterred by fear from eating it. In Germany, for instance, it has happened a number of times that hogs which have been condemned and ordered buried by the sanitary authorities have been dug up sur- reptitiously and eaten.-^ Trichinosis bears certain resemblances to typhoid fever and to acute tu- berculosis, but in addition is accompanied by oedema and intense pain. It arises from eating the infested meat in a raw or not well cooked condition. The trichinae are killed by exposure to 155° F., if they are not encapsulated; otherwise by a temperature of 158° to 160°. They are not affected by intense cold, putrefactive processes, nor ordinary smoking, but are killed l)y long ])ickling. The first symptoms ap})ear in a few days after ingestion, and indi- cate irritation of the alimentiiry canal. These are followed by febrile symptoms and intense muscular j)ains. Death may occur within a few weeks. In case of recovery, the parasites become encysted, and then arc incapable of producing further injury to their host. The manner in which they produce their effects is as follows : A\ hen the infested meat reaches the stomach, the digestive juices dissolve the capsules, and the parasites thus are left in a free state. In the intes- tine, they find conditions favorable to growth, and in a few days' time they grow so large that they can be seen Avith the naked eye and appear like fine threads. The female parasites when fully mature begin to produce young, each to the extent of upward of 500. These begin at once a migra- tion through the walls of the intestine and find their way to all parts of the body, and it is during this stage that the fever and intense pain are caused. In Plate I., Fig. 2, and Plate II., Fig. 1, are seen thin sections of muscle from a human subject, showing the worm coiled up and the 'Philadelphia Medical Journal, June 1, 1901. ^ .lournal of Medical Research, July, 1901, p. (54. 'For an account of such a case consult Zeitschrift fiir Fleisch- und Milclilivcicne, 1897, VII., p. 104. Trichinae in Human Muscle, showing Tliiekened Capsule. X 7S. Fig. 2. Triehinag in Pig Muscle. X 7S. 3IEAT AND FISH AND PARASITIC DISEASE. 41 thickened capsule formed about it. In Plate II., Fig. 2, it may be seen within the muscle of an infested pig. The sheep disease which is known as rot, which term, it must be said, is used to include a large number of abnormal conditions, but which, in its strict application, means a parasitic disease of the liver, is believed by many to be of sufficient importance to w^arrant the con- demnation of the flesli of the animal, but the scientific evidence on this point is to the effect that no possible harm can come to the consumer, even though the liver itself be eaten. The parasite infests not alone sheep, but cattle as well, and is known as the " fluke." Of the many varieties of flukes, there are but two kno%vn in the United States ; these are the common liver fluke (leberwurm, leberegel, schafegel, douve hepatique) and the large American fluke. The former infests cattle and sheep ; the latter, only cattle. The life history of the worm is exceedingly complicated, and is as follows : The adult or hermaphroditic worm fertilizes itself in the biliary passages of the liver, and produces an exceedingly large num- ber of eggs, W'hich pass to the intestine of the host with the bile, and are expelled in the feces. Such of the eggs as eventually reach water give rise after a longer or shorter period, according to temperature, to a ciliated embryo, which on its escape from the egg becomes a free swimming ciliated miracidium, and enters the body of certain species of snails, where it comes to rest. Here the organism grows, and, after a time, certain germ cells in its posterior portion develop a still differ- ent form of life, the ridise. These wander to the host's liver and in- crease in size, and in tiu-n develop from their germ cells the next gen- eration, w^hich are called cercarise. These leave the body of the snail and swim about in the water, and some become attached to blades of grass, where they form enveloping cysts and undergo anatomical changes. For their next stage it is necessary that they be received into the stomach of some herbivorous animal by being swallowed with the grass to ^vhich they are attached. On reaching the stomach, the cysts are destroyed and the parasites migrate to the liver and become adult hermaphrodites, thus completing the cycle. Occasionally they wander to the lungs and other parts of the body. In the liver, the parasites attach themselves to the walls of the bile ducts, wdiich may become completely blocked, and they cause the breaking-down of the surrounding tissues and general symptoms due to structural changes. The parasite cannot be transmitted directly from animal to man, since it requires an intermediate host, and in the stage preceding its final development it is not attached to material con- stituting human food. There are, to be sure, some instances of the dis- ease in man, though not by direct transmission from meat. The con- dition caused by flukes is known variously as rot, liver rot, fluke dis- ease, and distomatosis. 42 FOODS. TRANSMISSION OF DISEASE BY MEAT AND FISH. Living pathogenic bacteria in diseased meat and fish may gain access to the stomach in limited numbers and beget no disease. If they are not destroyed by the gastric juice, they have to contend with myriads of organisms normal to the intestines, and it is only when the condi- tions are such as to favor extensive multiplication that they are likely to produce harmful eifects. In meat that is cooked thoroughly, they are killed by the heat to which they are subjected. It is well known that the stomach has great jirotective power in its natural functions, since certain violent organic poisons may be taken into that organ without injury, while if (he same are intro- duced into the circulation, the results are fatal ; thus the venom of poi- sonous serpents is digested and made harmless by the stomach juices. An instance of the immunity conferred by cooking or by the process of digestion, or by both together, and of the fatal result of the admis- sion of the harmfid element of the same meat to the system through cuts and abrasions, is given by Lardier.' The case is as follows : A cow died suddenly of anthrax and was dressed for food. The meat was eaten by a large number of people, none of whom suffered the slightest inconvenience or injury. A number of cats, however, which ate some of the waste matters and licked tip the blood, died with some suddenness. A woman who bought the head and wounded her- self in the process of cutting it up had a charbon at the })lace of injuiy and died. Two men who helped skin the cow had charbon, but re- covered. A calf belonging to one of these two died of anthrax, and another man, removing the skin, ctit himself and died. The skin of the original nnimal was sold and the ])urehaser put it in a shed on his farm. Some time later, one of his cows died suddenly, and the man who dressed the carcass wotnided himself during the process, acquired a charbon, and died. How much farther this series of fatalities might have extended cannot be known, since the authorities took steps in the matter and prevented further fatalities. Manv instances are known in wliich the flesh of cattle dead of in- fectious disease has been eaten with im])unity. During the siege of Paris, for example, when the food supj^ly was exceedingly limited in amount, no one paid the slightest attentioji to the condition of meat in respect to disease, even glandcrcd horses finding a ivady market, and, so far as is knoAvn, no ill effects resulted. Many years ago, when the rinder/jcd was very prevalent in Bohemia, the diseased cattle were killed and buried by order of the government ; but the poorer class dug up the carcasses and cooked and ate them M'ithout suffering any evil conse<]uences Avhatsoever. During the prevalence of the same disease in England in the early sixties, the mcxit from the diseased animals in all stages of the distem- per was sent in enormous quantities to market, and sold and eaten without evil effects. A similar inununity has often been noticed after ^Kevue d'Hygiene, 1898, No. 5, p. 431, TBANSMISSION OF DISEASE BY MEAT AND FISH. 43 the consumption of the carcasses of animals dying from acute pleuro- pneumonia. In ordinaiy cases of this disease, which is peculiar to beef cattle, the effects are localised in the liuigs. Sometimes, in very pro- nounced cases, the flesh is altered in appearance, becoming dark and discolored, and it is also moist and flabby. It is believed that the meat is edible, if it possesses a normal appearance. The meat in rhiderpest undergoes no marked change in appearance, excepting in advanced cases, when it is dark in color and flabby and of disagreeable odor. In ordinary cases of foot and mouth disease, it appears that the car- cass is edible, but in exceptional cases, when the animal has suffered for a long time, the flesh may be so deteriorated as to be undesirable. As a rule, although the disease is very infectious, its course is mild and interferes only slightly, if at all, with the condition of the meat, which generally cannot be distinguished from that of healthy animals. Although many instances are known that show that the meat of animals suifering with anthrax may be eaten without injury, it is the unanimous opinion of those who have given the matter attention, that, no matter how good the meat may appear, it should be condemned and destroyed. If the meat is well cooked, accidents are rare, but many cases of fatal injury, involving a large number of victuns, have been traced to the use of such meat, presiunably not thoroughly cooked. In spite of the protection conferred by cooking, there is such an ele- ment of danger, even in the hanclhng of the meat, that its use should be discouraged and forbidden. In Scotland, it is a common practice with farm laborers and other poor to eat the meat of sheep which have died of acute inflammatory diseases, even of anthrax. The car- cass of an animal dying of disease is the perquisite of the herdsman and almost invariably is eaten after bemg salted. Xo j)recaution is taken, except to cut away the darker portions of the meat which show stagnation of the blood. Occasionally, serious consequences, due either to imperfect cooking or to insufficient salting, result from its consumption. It is held generally that the flesh of animals that have died from actinomycosis, puerperal fever, " strangles,^' hog cholera, and sheep scab is unfit for human food. Tuberculosis. — The cattle disease most commonly known in this country, if not elsewhere, is tuberculosis, and concerning the advisability of using the flesh of its victims, there is much difference of opinion, hei-e and abroad. The disease is more common in cows, especially those kept in confinement, than in steers and oxen. It is almost an unknown disease in the great herds roaming the western plains. In Berlin, in 1892-93, 15.1 per cent, of all cattle, 1.55 per cent, of swine, 0.11 per cent, of calves, and 0.004 per cent, of sheep slaughtered showed some evidence of the existence of the disease. In Copenhagen, in the years 1890-93, the figures were somewhat higher than those of Berlin, ex- cepting in the case of sheep. They were as follows : 17.7 per cent, of cattle, 15.3 per cent, of SA^une, 0.2 per cent, of calves, and only one sheep out of 337,014. At the abattoirs of Leipzig, m 1897, nearly 44 FOODS. half of the cows and about 20 per cent, of the other cattle were found to be tuberculous; 2.78 per cent, of swine, 0.2 per cent, of calves, and 8 sheep out of 49,559. Out of over 8,000 beeves of American origin landed, slaughtered, and examined at Hamburg, 4 were found to be tuberculous, while of the same number of native animals, 640, or 160 times as many, were found to be afflicted with the disease. At that time the German press had been carrying on one of its periodical agitations against the impor- tation of American beef cattle on account of the dangers to which native breeds were thereby subjected. In Great Britain, 30 per cent, of the cows are estmiated by Mac- Faydean to be tuberculous. In Belgium, of 20,850 animals tested with tuberculin in 1896, 48.88 per cent, reacted. In Denmark, of 67,263 thus tested, 32.80 per cent, reacted. In Mexico, about a third of the beeves slaughtered are tuljcrculous. In this countiy, while the percentage of affected animals is low, it is believed to be on the increase, both with cattle and swine. In tlie State of New York, it is said by veterinarians that, in some districts in which the herds are mainly of the hardy grades of the Ayrshire, Hol- stein, and Short-horn families, about 1 per cent, of the cows, and in others where Jerseys and Guernseys are more common, about 2 to 3 per cent, are tuberculous. In ^lassachusetts, those in a position to be best informed state that, among cows, the disease is much more frequent than in Xew York, but tliat it is rarely to be found in calves, steers, and oxen. In Pennsylvania, the State veterinarian believes that not over 2 per cent, of all cattle are tuberculous. At the large abattoirs of this country, about 1 in 2,000 cattle is found to be tuberculous. Dur- ing the two years ended June 30, 1899, 8,831,927 cattle were inspected by the Federal autliorities, and 7,015, or 1 in 1,259, were condemned on account of tuberculosis. During 1900, of 4,861,166 inspected, 5,279, or 1 in 921, were condemned. Of 23,336,884 hogs inspected, 5,440, or 1 in 4,290 were sufficiently affected to warrant condemnation of at least a part of the carcass. The organs involved most frequently in tul)erculosis of animals are the liver, lungs, kidneVs, brain, and udder. The muscles are affected very rarely, although the bacilli have been found in the expressed juice. At what stage of the disease meat becomes unfit for food, is a ques- tion over which there is much controversy. Extremists on the one side believe in condemning the entire carcass on the slightest evidence of disease in any part thereof, while those on the other side maintain that the entire animal may be used as food without injuiy. In Eng- land, the practice is to condemn any carcass in which the disease has made such extensive progress that the flesh lias become dete- riorated. The Royal Commission on Tuberculosis ^ concluded that meat from tuberculous animals may be consumed with imj^unity, if sufficient discrimination and care are exercised in slaughtering and dressing. Every part containing tubercles should be removed and ' The Veterinary Journal and Annuls of Comparative Pathology, June, 1895. TRANSMISSION OF DISEASE BY MEAT AND FISH. 45 destroyed, and the whole carcass itself in advanced or general tuberculosis. The French law excludes carcasses with generalized tuberculosis and those in which local lesions have involved the greater part of an organ. The same is true in Austria. In Prussia, the meat is held to be unfit for food if the animal has begun to show emaciation, but is passed as fit for human consumption if the disease occurs in only one organ, and in general, if the animal is well nourished. In Belgium, the law of September 30, 1895, permits the sale of meat of tuberculous animals after sterilization. Meat from tuberculous cattle is infective to other animals in very variable degrees. As a rule, the more advanced the disease, the more likely is the meat to be infective. Experiment has demonstrated that infection depends to a not inconsiderable extent upon contamination of the meat, in the process of dressing, by the hands, knives, or cloths, which have been in contact with tuberculous matter. Although lesions in the muscular tissue itself are not at all common, positive results have repeatedly been obtained in experunents in which the expressed juice of the meat has been injected into susceptible animals. Thus, Kastner obtained 9 positive results in 11 injections of the juice of the meat of 7 tuberculous animals, and Steinheil transmitted the disease to guinea-pigs by means of juice from meat apparently sound. Arloing inoculated the muscle juice of 10 tuberculous cows into guinea-pigs and demonstrated that that from 2 of the animals was mfec- tive. Nocarcl produced the disease with the muscle juice of but 1 of 21 tuberculous cows with which he experimented. All of these cows had been condemned at the abattoir on account of extensive lesions. Woodhead, Galtier, Humbert, and others have met with varymg degrees of success in similar experiments. That tuberculosis can be transmitted to animals by feeding them on tuberculous material has been abundantly j^roved, but the lesions pro- duced almost never involve the muscular apparatus, and many of the subjects escape infection altogether. It was reported, for example, by Thomassen, at the Tuberculosis Congress at Paris, that of 10 young pigs, each of which was made to eat 4.5 kilos of meat from animals with advanced general tuberculosis, but 2 were affected, and their por- tions had contained a quantity of splintered bone. Ravenel ' has held for a long time that food tuberculosis may appear first in the lungs and cervical glands, and cites the case of 2 cows which, fed on tuberculous material, developed extensive disease of the lungs and lesions nowhere else. As stated by Dr. D. E. Salmon ,2 Woodhead, St. Clair Thomp- son, and Lord Lister have shown " that infection through the mechum of the food may not necessarily be accompanied by disease of the in- testines. The organs first attacked after feeding on tubercular material may be the mesenteric glands and liver, or even the bronchial and mediastinal glands and the lungs." 1 Philadelphia Medical Journal, August 14, 1901, p. 284. '■' Bulletin No. 33, Bureau of Animal Industry, Washington, 1901. 46 FOODS. As is well knowD, the bacillus of human tuberculosis is fatal to many of the lower animals, but, as was pointed out by Professor Theobald Smith several years before Koch advanced the statement as original with himself at the Tuberculosis Congress at London in 1901, it fails to find in the bovine body the conditions necessary to its multiplication. There are, it is true, instances which show tl\at it may succeed in pro- ducing extensive lesions in the bovine subject, but there are, on the other hand, numerous experiments recorded which have yielded nega- tive results. It is said that, before the discovery of the specific organ- ism of the disease, Bollinger succeeded in producing tuberculosis in a calf with human tul)ercul(tus products. The exjieriments of Sheridan Delepine ' were equally successful. To one of two calves, he gave 50 c.c. of mixed sputum with its food. In 28 days it died. No evidence of tuberculosis was fi)uud in any organ except the glands connected with the alimentary canal, and these yielded virulent bacilli. The other calf Mas inoculated in the peritoneal cavity-, and 68 days after- ward gave a tuberculin reactit»n. Section showed marked tuberculosis of the peritoneum, extending to the pleura and pericardium. With the exception of a few lymph nodes connected ^ith the peritoneum, no other organs were affected. Salmon cites successful ex])erinients l)y Chauveau and Martin. The former fed emulsions made from tulier- eulous human lungs to two heifers and a bull, all under one year old, each receiving but two doses. All three became extensively infected. Martin fetl sputum to 6 calves, 2 of which remained healthy ; the othei's showed tubercular nodules in the intestines. In vieM' of these successful experiments and of the many negative results, it would be, as Salmon states, not entirely correct to say that human tuberculosis is not communicable to cattle, but that it is com- municated only with difficulty. Whether the bovine bacillus, by reason of its higher pathogenic power, may pass to the hiunan subject is, according to Smith,- a ques- tion concerning which we have, as yet, no definite knowledge. He has demonstrated that the bovine and human bacilli ])resent fairly pro- nounced differences in several res])ects, but concedes that, from clinical evidence at hand, it appears possible that bovine tuberculosis may be transmitted to children when very large numbers of bacilli are taken into the system in milk of cows with tubercular udders. Ravenel ^ believes that, from the evidence at hand, the bovine bacillus has a high degree of pathogenic power for man, especially in the earlier years of life. He has recorded * 3 cases of local infection from wounds in the finger, acquired in performing autopsies on tuberculous animals. Dr. L. Pfeiffer^ records the case of a veterinarian who wounded him- self in the left thumb while performing an autopsy on a tuberculous cow. The wound healed without pus formation. In the course of <> J British Medical Journal Oct. 2fi, 1901. ".Journal of Exporiiiicntal Medicine, \'ol. 3, Nos. 4 and 5. M^ancet, August 17. \W\. *Pliiladeli.liia Medical .Journal, Julv 21, 1900. ^Zeitschrift fur ilvgiene, 111. (1888), p. 209. TBANSMISSION OF DISEASE BY MEAT AND FISH. 47 mouths a nodule appeared iu the scar, and later, evideuce of iuvolve- nient of the lungs. A year and a half after the accident, he died of phthisis, and the joint was found to be extensively tubercular. Coucerning the possibility of transmission of tuberculosis by eating the meat of diseased animals, there is practically no evidence of value, but whatever danger there is, if any at all, is disposed of by thorough cooking, since thereby the bacillus is quickly killed. Siuce raw meat is frequently used as food, particularly in some diseased conditions, it is best, in order to be on the safe side, to see that meat so used shall be free from infective properties. Calves born of tuberculous cows are generally free from the dis- ease, and when not so, the mother has, in almost all cases, tubercular endometritis. MacFaydean reported a case of congenital tuberculosis before the Pathological Society of London, in May, 1899, in which the lesions were of the spleen, liver, and kidneys, with slight involvement of the lungs and almost universal invasion of the lymphatic glands. Of 3 other cases which had come under his observation since 1897, in one the cow had tuberculosis of the uterus, in another the disease was limited to the lungs, and iu the third no information as to the condition of the mother was obtainable. In the case of a cow with tubercular lesions in the lungs, liver, spleen, and udder, reported by Schroeder,^ small tubercles were found in the portal, mediastinal, and bronchial glands and in the spleen of the foetus. E,aveuel ^ records that a cow with advanced tuberculosis gave birth at seven months to a dead calf which showed no marked evidence of the disease, except that in the liver there were two white nodules about ^ inch in diameter, in which virulent bacilli were demonstrated. Later, when the cow was autopsied, the uterus was found to be free from tubercular changes, and the disease was found to be limited to the lungs and mesenteric glands. Fish are regarded commonly as incapable of becoming infected with tuberculosis, but one instance has been recorded in which the contrary was proved to be the case. Drs. Dubar, Bataillon, and Terre commu- . nicated to the French Academy of JNIedicine the details of this case, which are as follows : The sputa and discharges of a woman suffering from advanced tuberculosis of the lung and of the intestines were regularly thrown into a pond, in which were large numbers of carp. After a time, many of the fish were found to be dead, and, on exam- ination, their livers and other organs were found to be full of tubercle bacilli. Healthy fish, fed experimentally on this and similar material, became tuberculous within a few weeks, but the infected fish were proved to be quite devoid of power to infect warm-blooded animals, and, therefore, it appears that there can be little danger from the con- sumption of tuberculous fish. A similar experiment conducted by ' Hermann and Morgenroth^ gave negative results. They fed tuber- 1 Zeitschrift fiir Fleisch- und Milch-hygiene, 1900, p. 79. ^ Proceedings of the Pathological Society of Philadelphia, 1899. ** Hygienische Rundschau, 1899, p. 857. 48 FOODS. culous sputum to goldfish, which ate it eagerly and for weeks eliminated living tubercle bacilli in their feces, but they themselves remained per- fectly healthy. The danger from other animal diseases will be considered under meat poisoning. Typhoid Fever and Cholera. — Foods of all kind may be made the bearers of infection to man, though themselves in good condition. Par- ticularly is this noticeable with regard to oysters, which have many times conveyed the specific organisms of typhoid fever and cholera. In 1880, Sir Charles Cameron* brought to the attention of the pro- fession, that oysters, transplanted from the coast of the County of A\ ex- ford to tlie northern shore of Dublin Bay, had for some years been much subject to disease and had died in large numbers. Specimens which were examined were found to contain sewage matters, and inves- tigation showed that the beds " were literally bathed in sewage." He offered the suggestion that raw oysters, taken from the shore close to sewer outlets, were, perhaps, as likely to act as the vehicle of typhoid fever and other diseases as contaminated water or milk, and advised that " oyster beds should not be laid down at any point on or close to the mouth of a sewer." But the warning appears to have excited no more than a languid mterest until 1893, when the late Sir R. Thorne-Thornc, in a report to the Local Government Board, stated his belief that the sporadic cases of cholera which had occurred at various inland places in England in that yenr were due to oysters and other shellfish from sewage-contaminated water at Grimsby, where there had been a small outbreak of tlie disease. In the following year occurred the well-known outbreak of typhoid fever at Wesleyan University, which was so ably and conclusively traced by Professor Conn ^ of that institution to polluted oysters. On October 20, 1894, several of the students were reported as slightly ill, with a moderate degree of fever. The number of cases grew from day to day, and shortly included several of undoubted typhoid fever. By November 1st, there were 20 cases of the disease, Avhich number was shortly further increased to 23. All of the victims were men ; there was no illness among the 58 women students. Investi- gation completely absolved the water supply, the general and par- ticular food supplies of the various boardhig places, and the local conditions of the dormitories and outside lodgings of all suspicion of blame. It appeared that nearly all of the victims were members of three of the seven college fraternities. The combmcd membership of the three was about one hundred. On October 1 2th, eight days before the development of the first symptoms, all seven fraternities had had their initiation ceremonies and had celebrated in the usual way with a supper. Investigation of the origin of the components of the su[)])ers showed that there was but one disli iVom a common source, and that was oysters. The tliree afilicted societies and one otlier liad obtained ' British Medical .Journal, September 18, 1880, p. 471. =* Medical Record, Dec. lo, 189-1, p. 743. TEANS3IISSI0N OF DISEASE BY MEAT AND FISH. 49 their oysters from a local dealer ; of the remaining three, two had had no oysters, and the third had had some from a dealer in Hartford. Of the four supplied by the local dealer, one had eaten the oysters cooked, and its membership was not invaded. Thus the trouble was sifted down to the raw oysters from the local dealer. But there was one victim who was a non-society man, and, clearly, his case could not be traced to the initiation supper. Investigation of his dietetic history established the guilt of the local oyster supply even more securely, for it was shown that he had eaten raw oysters from the same lot at the shop of the dealer. It was learned, too, that 5 men from Yale had attended the exercises of the societies in which the outbreak occurred, and inquiry developed the information that 2 of the 5 were seized with typhoid fever some weeks after their return to New Haven. Further investigation revealed the fact that the incriminated oysters had been brought from a bed in Long Island Sound, and, on October 10th, two days before use, had been stored in a bed at the mouth of the Quin- nipiac River, a short distance (300 feet) from the outlet of a private drain from a dwelling, in which 2 persons lay ill with typhoid fever. Shortly after the publication of this case. Sir William Broadbent ^ published the facts of a number of cases of fever, seen by him in consultation, which appeared to be connected with the ingestion of raw oysters, although no absolute proof was adduced. These included the following : (1) A young woman who, during convalescence from child- birth in a sanitarily perfect house, in which no other inmate was sick in any way, had eaten raw oysters. Ten days afterward, she came down with typhoid fever of an unusually severe type, from which she did not recover. All the water and milk which she had taken had been boiled. (2) Two young men, inmates of the same house, in which there was no other sickness. Both the house and the place of business where they were employed were in good sanitary condition. They were seized, simultaneously, with typhoid fever of an unusually severe form, for which there was no apparent cause, except that, ten days before the appearance of symptoms, they ate an oyster supper together. Both cases terminated fatally. (3) A young woman, who, ten to fourteen days before, had on two occasions eaten a half-dozen oysters with a cousin of the same age. She had a mild attack of the dis- ease, as did also her cousin, who, in the meantime, had gone to Italy, where she, too, was seized. (4) A man and wife, who were seized with the disease at the same time in a house which was sanitarily per- fect. No possible cause was apparent other than that, 2 weeks before, they had indulged in oysters. (5) A young man, who had been sick for 3 weeks with influenza and bronchial catarrh. He had partaken freely of oysters, and had developed typhoid fever. No other person in the household, which was a large one, had any sickness. (6) A clergyman and his daughter, living in a rural district where typhoid fever was unknown. The sanitary condition of the premises was per- fect, and no other member of the household was sick. Inquiry ^ British Medical Journal, Jan. 12, 1895, p. 61. 50 FOODS. showed that, about 2 weeks before, they had twice had oysters from Loudon, aud that they alone had eateu them. Shortly after the report of Broadbeut's cases, N. J, Johnsou-Lavis ^ related certain facts which he had noticed while in practice at Naples in 1879. Gastro-intestiual disorders were especially prevalent among strangers, varying in intensity from evidence of simple irritation to the most severe forms of typhoid fever. Whenever he aud his wife ate oysters, they suffered from colic, diarrhoea, aud tenesmus. Gastro- intestinal symptoms in his patients very commonly followed eating raw oysters. These cases included chronic gastro-intestiual disturbances, very stubborn in character, aud typhoid fever, often of a very severe type. He noticed, too, that though no sickness was caused by eating oysters at the several places along the Italian coast from which the Naples supply was obtained, when they were brought to Xa})les and kept for weeks and sometimes mouths in the harbor in a bed less than 60 feet removed from the outlet of one of the main sewers, their use was by no means unattended by risk. At this place, individual dealers stored their oysters in baskets, which were pulled up through the filthy water as occasion demanded. Some of the oysters were examined. They yielded evidence of sewage in the M'ater between their shells. Arthur Newsholme,' Medical Officer of Brighton, England, re- ported that, during 1 894, 83 cases were returned to him as typhoid fever. In 15 instances, the original diagnosis was found to be incor- rect, and an equal number were found to be imported cases. He in- vestigated the probable causes of the remaining 53 cases, and decided that no less than 1 5 were due to oysters, aud 6 to other contaminated shellfish (clams, cockles, and mussels). In other words, 40 per cent, of the genuine cases of ty})hoid fever were due to these articles of food. In a later connnunication,' after a thorough examination of the cases occurring during a period of four years, he reported the percentages of cases probably due to ovsters and mussels as follows : In 1894, 38.2 ; 1895,33.9; 1896,31.8; 1897, 30.7. Chantemcsse * relates the following case : There had been no case of typhoid fever in the village of I'Herault Saint Andre de San- gonis for about a year, when, on February 15th, a shopkeeper received a consignment of oysters from Cette. The entire lot was consumed by 14 persons, all of whom were made sick. In the 6 dwellings in which the victims lived, no other inmates were sick in any way. Eight of the number were made only slightly ill, the symptoms, which in- cluded abdominal })ain, vomiting, diarrhcea, borborygmus, anorexia, and general malaise, lasting but 2 or 3 days. The 4 youngest, who ate but a few, were very sick for a much longer time (15 to 25 days), but recovered. The stools were very offensive, were passed with })ain, and were dysenteric in appearance ; there was tym])auites with teuder- 1 British Medical Journal, March 9, 1895, p. 559. '' Tl>ideni, .Tunc 8, 1895, p. 1285. « Puhlic Iloaltii, Sei)tcnihcr, 1898. * Bulletin de 1' Academic de Mcdecine, 1890, 35-36, p. 588. TRANSMISSION OF DISEASE BY MEAT AND FISH. 51 ness and gurgling. All 4 were greatly prostrated. The remaiuing 2, a woman of twenty and a man of twenty-one, developed very severe cases of typhoid fever. The woman died. To satisfy himself as to the probability of oyster infection, Chante- messe secured specimens from several sources (Marenne, Ostend, Portugal, etc.) and made bacteriological examinations. They yielded an abundance of bacteria, and many were found to contain B. coll com- munis. He placed some of them in water intentionally infected with typhoid stools and cultures, and after 24 hours, removed them and kept them another like period before subjecting them to bacteriological test. They yielded the typhoid organisms and B. coll communis in great numbers. Mosny,^ to whom the French authorities referred the whole subject of mollusk poisoning for investigation, has reported that 5 members of a family of 7, living in a village in a suburb of Paris, in which there had been no case of typhoid fever in 4 years, were made sick after eating oysters sent to them from Cette. Four were seized in the evening of the following day mth gastro-intestinal disturbance, which lasted 24 hours. On the eighteenth day, a youth of 17 years devel- oped unmistakable symptoms of typhoid fever, of which, 9 days later,, he died. In March, 1897, Chatin^ reported the case of a family, of which several members were stricken with typhoid fever after eating- oysters from a bed which was contaminated by sewage. In 1889, De Giaxa^ made a series of investigations of the influence of sea water ou pathogenic bacteria, and found the followmg conditions to be favorable to growth and midtiplication of micro-organisms m harbor water : Shallowness, stagnation and high temperature of sur- face water, abundance of vegetation, and admixture of sewage rich in.' organic matter. Certain pathogenic forms were found to succumb very quickly to the influence of the ordinary species of water bacteria and others to be far less susceptible. Some were found to thrive well in sterilized sea water and to retain virulence for many days. It was sho^ATi by Foote,^ after the outbreak at Wesleyan University^ that typhoid cultm-es, introduced within the shells of oysters from the bed from Avhich the incriminated oysters were derived, were virulent at the end of 48 hours, which was the period which elapsed between the gathering and consumption of those which caused the outbreak. Fur- thermore, it was demonstrated that, if the specimens were kept at 57° F., the organisms were active as long as a month later. Joseph Polak,^ of Warsaw, examined oysters from Ostend, Hol- land, and Odessa, and concluded that, during transportation, the life processes have an undoubted inimical influence on bacteria, diminishing, and in certain cases destroying them completely. His conclusions were distinctly opposed to those of others who had determined that the ty- ^ Eevue d' Hygiene, Jan., Feb., and March, 1900. ^ La Semaine medicale, 1897, p. 9. "* Zeitschrift fiir Hvgiene, VI., p. 162. * Medical News, March 23, 1895. * Sanitary Record, April 30, 1897, Supplement, p. 47. 52 FOODS. phoid fever organism lives longer in the tissues and juice of the oyster than in sea Avater itself. Klein detected B. coll communis in oysters from typhoid-mfected beds, and found that oysters, kept for a time in sea water inten- tionally infected with B. fyjAosus, yield the organism after 4 to 18 days. Cholera bacteria were demonstrated in an active state after 4 to 18 days. Wood subjected oysters to cholera-infected sea water, and found the bacteria at the end of 18, but not after 20, days. According to Boyce, the typhoid orgtmism will not grow in the tissues of the oyster, and, in fact, perishes rapidly therein, if the oysters are removed to pure sea water. But, according to Klein and Foote, the organism lives, multiplies, and is virulent for a long time, if the oysters are stored in polluted water. Klein found it virulent after 3 weeks, and Foote reported that, during the first 2 weeks of immersion in typhoid-infected water, it multiplies in the tissues of the oyster and then diminishes, but can still be found after 30 days. From a series of experiments undertaken to determine the question of viability of the typhoid organism in sea water and "within the oyster, Bordoui-Uffreduzzi and Zernoni ' concluded that it will live over 2 weeks in sea water and from 3 to 4 days in oysters, without lessening of virulence. Oysters from Spezia, Venice, and elsewhere, were exammal to determine the presence of the tA'jjhoid organism in the water contained between the shells or in the tissues. The results were negative on this point, but the colon bacillus Avas isolated from oysters from 3 different sources. Oysters immersed in sterilized sea water, which later was infected with cultures of the typhoid organism, vielded virulent bacilli from the water between their shells up to the ninth day of examination, but never from the tissues themselves. Other observ^ers have found the bacteria of cholera and typhoid fever, B. coli communis, B. proteus vulgaris, and other organisms, in oysters ervisi(m of oyster beds. In the investigation of outbreaks of typhoid fever supposedly due to oysters, bacteriological proof of specific infection of those eaten or of others from the same lot always has been and always will be wanting, since, long liefore the appearance of the first symptoms of the disease, the material is no longer available for investigation. But, in view of the fact that pathogenic bacteria have been found in the water betAvcen the shells of oysters from jwlluted beds , that they have been known to live for days in the tissues and retained Avater ; and that, in the cases uiA^estigated, the beds have been found to be exposed to the influ- ence of sewage, we may, therefore, ])roperly conclude that a Oausal rela- tion is A'eiy possible. The danger of infection arises Avholly from the presence of sewage in the AA'ater Avhere the oysters are planted or stored. The remedy lies either in transferring the beds to cleaner situations or in storing the con- ' Giomale dellu Reale Societri Italiana digiene, 1899, p 500. POISONING BY MEAT AND FISH. ' 53 taminated oysters in clean sea water nntil tlie bacteria either have per- ished or have been washed away. AVhat constitutes a sufficient length of time to insure purification, is a matter of some disagreement. Many believe that a week is enough ; others, that 16 days should be allowed. Ovsters should not be stored where sewage matters can reach them through long distances by ciu'rents along the shore, nor where prevailing winds can exert a harmful influence to the same end. Poisoning by Meat and. Fish. Animal foods are the frequent cause of most distressing disorders which not rarely have a fatal termination. Some of these are due to poisonous properties inherent in the living animal, some to bacterial poisons formed in meats showing no evidence of unwholesomeness, and some to decomposition products developed during storage or puti'e- faction. 1. Poisoning Due to Substances Normally Present in the Living Organism. — As has been stated, certain species of fish are always poi- sonous and others only at times, and in some cases only individual members are so constituted. Certain species are so well known to be poisonous in perfectly fresh condition that they never are eaten by the natives of the places where they are found, except for purposes of sui- cide. Some have poisonous glands connected with their fins, some have poisonous ovaries, and others are poisonous throughout. Some are poisonous only in the raw state, and others whether cooked or not. The symptoms produced vary widely, sometimes mdicating gastro- enteritis, sometimes involvement of the central nervous system. The mussel is regarded not uncommonly as an intrinsically poisonous shellfish, but the weight of evidence indicates that mussel-poisoning is due to conditions of disease or infection arising from residence in pol- luted water. Its poisonous properties have long l^een recognized, and have been the subject of a number of dissertations by early writers ; thus Behrens, De affectionibus a comestis mytiiUs, Hannover, 1735, and jSIoehring, Jlytidorum quonmdam venenum et ab eo natas papidas cuti- culares Epistola, Xuremberg, 1744. In France, where great cjuantities of mussels are eaten, cases of poisoning thereform are rare, owing doubtless to the fact that those taken from polluted harbors are kept for a week or more in clean water elsewhere. 2. Poisoning Due to Bacterial Products in Meats and Fish. — What is known commonly as meat-poisoning, fish-poisoning, and sausage-poisoning is due to the products of a number of micro-organisms having no connection with the usual diseases of man. These products, which include toxins and ptomains, cause an extremely wide variety of symptoms, which, as may be observed on examination of the collection of reported outbreaks given below, indicate the possible derangement of function of practically every part of the system. There are two groups of symptoms, however, ^vhich are fairly constant, either one of which may predominate over all the rest. These are (1) the manifes- 54 FOODS. tations of profound disturbance of the gastro-intestiual canal, and (2) those indicating more or less intense poisoning of the central nervous system. Prominent among these latter are impaired vision (dilated pupils, ptosis, amphodiplopia, etc.), and glosso-phaiyngeal })aralysis ; and when those are present, the case is said to be one of "botulism." This term, which came into existence by reason of the fact that many of the earlier observed cases of food-poisoning were traced to sausages (/)ot(ilus, a sausage), is, in the light of our present knowledge, unfortu- nate and misleading, for the condition may be caused not only by sausage, but by any form of meat and fish which may happen to be contaminated by the micro-organisms which produce the peculiar toxin (or toxins) by which the manifestations are caused. And it is not tme, as is supposed In* some, that botulism is caused by the proteid bacterial jDoisons alone (commonly known as toxins), but by certain of the basic crystallme products of decomposition, Icnowu as ptomains, as, for example, mytilotoxin, a ptomaiu isolated by Salkowski and Brieger from contaminated mussels. Xot uncommoidy, ptomains are regarded as necessarily jioisonous substances. This, however, is far from being the tnith. They are products of decomposition brought about by micro-organisms which break up the complex organic matters into less complex compounds, "which in turn are split u\) into pnidiiets of diminishing comjilexitv, until the final products are water, hydrogen, carbonic acid, sulphur- etted hydrogen, ammonia, nitrogen, and salts. During this process of decomposition, at different stages, the ptomains, which are organic bases, are formed. Some are poisonous, but the great majority of those thus far isolated are wholly inert. All contain nitrogen, but not all contain oxygen, thus resembling the vegetable alkaloids. The variety of ptomains formed depends upon the kinds of micro-organisms at work, the nature of the substance undergoing decomposition, and the conditions of temperature, access of air, and other attendant circum- stances. One species of bacteria may produce no ptomains from one kind of material, and poisonous or inert ones from another. At one stage of decomposition no ptomains may be formed, at another several may be present, and later these may have disappeared completely, for they are but intermediate products. Brieger has isolated a nnmber of varieties of ptomains from decom- posing meats and fish, including neurine, choline, and one which ajipears to be identical with muscarine (all three of these are antagonized in their p(>isonous action by atropine), and neuridine, putrescine, cadavcrine, another which produces efix'Cts similar to those of curare, and others. Vaughan discovered the very important })tomain, tyrotoxicon, in milk and cheese. Many of the poisonous compounds formed during putrefaction retain their active character long after the org-anisms through whose agency they have been produced have perished. This was noted as early as ISoT) by Panum, who found that the poison of certain ])utrid meat retained its activity even after it had been boiled 11 hours, and his POISONING BY MEAT AND FISH. 55 observation has repeatedly been confirmed by others. Xatui'ally, no amount of cooking will suffice to render such meat harmless. The physiological action of these poisons is widely different. Some cause intense gastro-intestinal irritation, some act directly on the heart, some on the central nen^ous system, and some on particular centers. Very different effects are produced in different people, owing perhaps to varying degrees of susceptibility and also to unequal distribution of the poison through the mass of meat. The extent to which the putrefactive process has advanced is by no means of such importance in the determination of the question of pos- sible ill effects, as the nature of the engaged bacteria and of their products, for meat may be extremely putrid and yet not be poisonous, and, on the other hand, may be apparently normal and yet deadly in its effects. Many savage peoples prefer putrid fish and meat, and the more rotten it is, the greater their enjoyment in its consumption. In less degree, the same is true of many of the most enlightened people, who prefer game when decomposition is fairly well advanced. On the other hand, the severest outbreaks of food-poisoning have followed the eating of meat apparently not undergoing decomposition. Indeed, the majority of persons will reject meat which has the slightest taste or odor indicating beginning putrefaction, since even this makes it repug- nant to the senses. In many cases, the poisonous principles appear to be developed after the meat has been eaten, through changes occurring within the intestines. The bacteria which have thus far been shown to have been the cause of outbreaks of meat- and fish-poisoning include certain spore-bearing anaerobes isolated by Van Ermengem (jB. botuUnus), and Klein (^B. enteritidis sporogenes), a number of derivatives of B. coli isolated by Gaertner (5. enteritidis), Basenau {B. bovis morbijlcans), Kaensche (-B. Bredaviensis and B. Jlorseelenensis), Gaffky and Paak [B. Friede- bergensis), Abel, Glinther, and others, besides B. proteus vulgaris, B. proteus fluorescens, B. proteus mirabilis, B. pAscicidus agilis (Sieber), B. liydrophilus fuscus (Sanarelh), a micrococcus (Vaughan), and others unnamed. The first-mentioned {^B. botuUnus) produces an extraordi- narily virulent toxin, which has been the subject of carefol investi- gation. It was isolated from cultures of the bacillus, supplied by the discoverer, by Brieger and Kempner,^ who proved it to be related closely to the toxins of diphtheria and tetanus, from which it cliifers in the important respect that it can affect the system through the mucous membranes. Xext, Kempner ^ investigated the subject of immunity to the toxin, employing bouillon cultures killed by the application of toluol, culture filtrates free from bacteria, and the concentrated and purified poison, the strength of which was determined accurately ^^^th guinea-pigs. The first experiments ui immunizing guinea-pigs and rabbits proved that with them immunity cannot be attained even when beginning with the smallest possible dose, since in every case, after a shorter or longer ^ Deutsche medicinische Wochenschrift, 1897, Xo. 33. '^ Zeitschrift fiir Hygiene und Infectionskrankheiten, XXYI., p. 481. 56 FOODS. interval, the animal perished. With goati?, it Mas found that immunity can be conferred by repeated increasing subcutaneous injection, and that the serum of the immunized animal possesses a very high protective power, as is shown by the fact that protection is secured by injection performed 30 hours before the introduction of the poison. It was found also that the antitoxin would save guinea-pigs when administered 24 hours after inoculation with a dose which would be fatal ordinarily in 48, and even when decided clinical signs of poisoning were already present. Later, Kempner and Schepilewsky ^ began a research on the possible affinity of nerve-substance for the toxm, Avhich, as shown by clinical symptoms and jiathological examination, exhibits a decided affinity for certain parts of the central nervous system. The test poison was so standardized that 0.000005 cc. represented twice the dose necessary to kill white mice of 15 grammes weight in from 2 to 3 days. The brains of recently killed guinea-pigs were rubbed Avith physiological salt solution in the ratio of 3.3 grammes to 10 cc, and their cords were treated m a similar manner. Both emulsions were practically neutral in reaction. Pieces of the liver, kidney, spleen, muscles, and marrow were treated i:i the same way, to be used for comparison. In the first series of experiments, 1 cc. of the brain or cord enudsion, mixed Avith three to four times the fatal dose of the toxin, was injected under the skin of a number of mice, and, as controls, other mice were injected with the same amount of toxin alone, and still others with the toxin mixed with emulsions of other organs. Tlie results showed that the brain and cord exert a decided preventive and curative influence, and that the other organs do not. A^'ith mixed injection it was always possible with 1 cc. of the emulsion to counteract 3 times the fatal dose, while with 4 times the dose about half the animals died. With separate, but simultaneous, injections, only about half of the animals survived twice the fatal dose, the others dying about as quickly as the controls. As a curative agent in cases where the poison had been exhibited 6 and 12 hours previously, the emulsion did not give such favorable results, for those treated after 12 hours died as soon as the controls, and the others lived but a day longer. Whatever the protecting substance may be, it was proved that it is in combination with nerve-substance and insoluble in Avater ; its influ- ence is altered materially by high temperatures, but not by keei)ing several days in ice. Milk, butter, yolk of egg, and other animal flits were tried in the same way, but no results were obtained, excepting with butter, with Avhich two guinea-pigs were protected. Even this failed Avith other animals. A thin oil emulsion, mixed Avith 2 and CA'en 4 times the fatal dose, gaA'e positive results. Lecithin and cholesterin, substances normally present in nerve-substance, proved to haA'e antitoxic jioAver, Avhich Avas unimpaired by boiling or heating ; but large amounts Avere necessary. Other substances, as cerebrin, nuclein, and bile, proved to ' Zeitschrift fiir Hygiene unci Infcctionskrankheiten, XVII., p. 213. POISOyiNG BY MEAT AND FISH. 57 be inert, but antipyrin in larger amounts than 0.10 gramme (0,15 to 0.20) sufficed to neutralize 0.00003 cc. of the toxiu. Given iu doses of 0.10 gramme, antipyrin had no eifect, the animals dying at the same time as the controls, but when the animals were treated with antipyrin as a preliminary measure the above dose was efficient. Onset and Course of Symptoms. — The first symptoms in cases of poisoning by fish and meats may occur within an hour or two after eat- ing or may be delayed a number of days. In one outbreak cited (see Poisoning by Herrings, page 61), m which 5 persons were seized, the initial symptoms appeared m 2, 3, 5, 7, and 9 days respectively ; ordinarily they appear within a few hours — 3, 6, 12. AVhen numbers of persons are affected by the same food, the onset is by no means uni- form. In the Ellezelles case (see page 69), in which 20 persons were seized, the time iu which the symptoms first were manifested ranged from 3 to 36 hours, but as a rule, it is the appearance within the same day of similar symptoms in a number of persons which calls attention to the food supply as a common cause of the trouble. Poisoning by ptomains is manifested generally within a few hoiu's. In cases of rapid onset, the progress either to recovery or a fatal termination is commonly short, but may be sometimes a matter of months, and in these exceptional cases eventual recovery is probable. The shortest case on record is that of mussel-poisoning at Wilhelms- haven (see page 60), in which 1 victim died in 2, another in 3, and 2 others in 5 hours after eatina:. A peculiar tendency to relapses often is observed. The patient begins to improve, when suddenly the original symptoms reappear with ecjual, greater, or climinLshed intensity. Improvement may be succeeded again by a relapse, and the alternation may obtain for many months. The toxins secreted by the original invading bacteria are antagonized by antitoxins produced by the system and improvement occurs ; then dur- ing this interval the spore-bearers find opportunity to develop a new crop of bacteria, which, again producing toxins, cause a recurrence of the original symptoms. Nature of Symptoms. — As has been stated, the effects produced vary very greatly, but the symptoms of abdominal disturbance and of poi- soning of the central nervous system are the most constant as well as most predominant. Fever may or may not be present ; usually it is not, but in some outbreaks temperatures exceeding 104° F. have been recorded. In some cases, the temperature is subnormal. Disturbance of the circulation is more common than fever, the pulse being small and rapid, and sometimes dicrotic. In a few instances, marked embar- rassment of respiration has been noted. In most of the recorded cases, no mention is made of mvolvemeut of the kidneys, but in some in- stances evidence of acute nephritis has been observed. Dysuria, anuria, and paralysis of the bladder are not unconmion. In most cases, extreme muscular weakness is a prominent symptom, and not infre- quently muscular pains and cramps. While diarrhoea, long continued, IS a most common occurrence, in many cases most obstinate constipation. 58 FOODS. sometimes following diarrhoea and sometimes present from the first, is noted. In some cases abdominal symptoms are bv no means prominent, and in others they are practically the only ones observed. The symp- toms of involvement of the nervous system include those mentioned above, and drowsiness or insomnia, headache, dizziness, delirium, dimin- ished co-ordination of movement, numbness, cramps, convulsions, and paralyses. Post-mortem Appearances. — The post-mortem appearances observed in cases of poisoning are very inconstant, both as to extent and kind, and are by no means proportionate to the severity- of the symptoms. Even wlien a number of individuals succumb to the same influences, the appearances may show but little in common. Thus, in the Welbeck case (page 66), one showed nothing more than a few bright red patches in the stomach ; a second, congestion of the gastro-intestinal mucous membrane; and a third, severe parenchvmatinis inflammation with dis- tention and plugging of the arterioles and capillaries of the Malpighian corpuscles by emboli of bacteria. The most extensive changes observed are those occurring in poisoning by mussels and oysters, in which cases the extremely ra}>id onset and the veiy short course to a fatal termina- tion suggest the action of jioisonous ptomains. Indeed, animal exjieri- mentation has demonstrated that certain of these comjiounds produce these verv changes, which include great enlargement of the spleen, pimc- tiform ecchymoses and hemorrhagic infarctions, and fatty degenera- tion of the heart, liver, and kidneys. In cases of meat-poisoning, the appearances noted range from a few red patches in the intestines to severe gastro-enteritis with destructive changes in all the principal viscera. Character of Meats which Cause Poisoning. — In general, outbreaks of poisoning iwv caused by the meat of animals slaughtered while suffer- ing from diseases other than those which are best known to the public because of the grejtt destruction wrought when raging in epidemic fomi ; but they may also be traced to the flesh of perfectly healthy animals which has become contaminated, both in the raw and cooked states, by poison-producing bacteria. The most dangerous fu'ms of meat-poisoning are those due to the pytemias, septicaemias, and pneunio-enteritis, and the greatest intensity of action is produced by preparations made from the entrails. In a majoritv of the rejwrted outbreaks, the meat has been consumed eitiier raw or oidy impci'fectly cooked, or after being kept a day or two after being cooked. The meats most commonly the cause are pork and its preparations, and veal. Both yield a considerable amount of gelatin, and this fact has been suggested as having an important bearing, since this material is a medium which offers favorable opportunities for the growth of bacteria. Most of the reported outbreaks have occurred in tlie countries of Europe, where the meat sup]ily, in consequence of being veiy restricted, is utilized to its fullest extent. Viscera which with us are rejected as refuse, and the flesh and viscera of animals slaughtered in consequence I POISONING BY MEAT AND FISH. 59 of sickness, with the consent and approval of ofl&cial veterinarians, are sold and eaten. Another reason for the frequency of the outbreaks is a very common preference for scraped or minced raw meats and for sausages of domestic manufacture made under most unsanitary con- ditions. Veal. — According to Vallin, in a communication to the Academy of Medicine in 1895, a large number of outbreaks of poisoning in Germany, Switzerland, and elsewhere are due to the consumption of veal from animals either sick or too immature. Darde and Drouineau ^ relate that they have seen nearly the whole strength of a military company, 135 out of 147, poisoned by eating roast veal. The symp- toms appear generally in from 6 to 24 hours, and include vomiting, purging, and great prostration. Dilatation of the pupil is common, but not constant. Occasionally, skin eruptions appear. By Vallin,^ and by others as well, it is deemed probable that veal- poisoning is due largely to the existence of septic pyaemia and septic pneumo-enteritis in calves, and Van Ermengem has suggested that a number of septic diseases of these animals are grouped commonly under the head of diarrhoea. He fed the fresh meat of one of these calves to mice and guinea-pigs, which died "within a few days with enteritis. From the bone marrow he isolated an organism which appears to be re- lated closely to Gaertner's B. enteritidis, and which on inoculation into animals produces a fatal enteritis. Beef. — Beef-poisoning has been noticed with considerable frequency, following the use of meat from animals slaughtered while sick, and it has been pointed out by several observers that certain septic diseases of cattle are especially prone to render meat poisonous. These include the septic form of calf paralysis, hemorrhagic enteritis of calves, septic metritis of cows, various intestinal disorders, the septico-pysemic dis- eases, and a number of others. Gaertner's B. enteritidis was discovered by him originally in the flesh of a cow that had been slaughtered on account of a severe diarrhoea, and in the spleen of a person who died in consequence of eating it. He showed that not only the bacillus, but also its boiled bouillon cultures, are highly toxic. Many deaths have been recorded as a consequence of eating the cooked meat of cows slaughtered on account of puerperal fever, and it was from such an animal that Basenau isolated B. bovis morhificans. This cow showed such lesions of the viscera that the director of the Amsterdam abattoir forbade the use of the meat. Basenau ^ has examined the flesh of beeves which had succumbed to a variety of diseases, and he has isolated a number of species of bac- teria bearing a close resemblance to B. bovis morbificans, all of which are fatal to mice. Some of them produce poisonous matters which withstand boiling without impairment of their properties. Ordinary inspection being useless for determining whether such meat is infected, ^ Archives de Medecine et de Phaiinacie militaires, 1895. ^ Revue d'Hygiene, 1895, XVII., p. 473. ^ Archiv fiir Hygiene, XXXII., p. 219. 60 FOODS. he recommends that bacteriological and feeding experiments slionld be instituted tooether within 24 hours after slauohterino-. If no colonies are observed at the end of 24 hours and no bacteria are seen in the tis- sues, the meat may be regarded as safe to eat. If colonies are yielded, the acceptance or rejection of the meat must depend upon the results of the feeding experiments. If the mice fed on the raw meat die and those fed on the cooked meat survive, it may be concluded that the meat is safe, if thoroughly cooked. If both die, the meat should unhesitatingly be condemned. Sausage. — Sausage has long been recognized as a very common cause of poisonmg, and has a much larger record of accidents than any other meat or meat compound. This is due in large part to a very common practice of making use of all manner of uninviting fragments and scraps of meat, offal, and the flesh of sick and ill-conditioned animals in preparing sausage meat, and perhaps to a greater extent to the extremely unsanitaiy methods of manufacture which obtain in those districts where this form of poisonmg is most prevalent. In many instances, the symptoms caused are due to the presence of ptomains, and in niany to the contained bacteria and their toxins. In most instances, it is impossible to fix the blame upon any in- dividual constituent, nor aside from its scientific interest is this of special importance. The symptoms present as Avide variations in char- acter as are observed in any other form of food-poisoning The process of smoking, to which certain varieties of sausages are subjected, while not destructive to the bacteria of putrefaction, is often successful in masking any unjjleasant smell or taste due to change. Cases Illustrative of Poisoning by Fish and Meat. Poisoning by Mussels. Case I. — At Wilhelmshaveu, in 1885, several longshoremen and their families, 19 persons in all, were stricken with very severe symptoms shortly after eating a meal of mussels. The symptoms were in general the same in all, regardless of the amount eaten, and included nausea and vomiting without abdominal pain or j)urging, trembling, constriction of the throat, dizziness, and diminished coordination of movement similar to that due to alcoholic intoxication. There was no fever. Speech was difficult and thick, and in a short time the legs were miable to support the body. The pupils were dilated and unresponsive to reaction tests. The extremities Avere cold and numb. Four deaths occurred, one within tMo hours, one in three and a-half, and the others within five hours from the time of ingestion. The autoj)sy in the only case examined revealed enteritis, enormous enlarge- ment of the spleen, numerous hjemorrhagic infiirctions, and fatty degen- eration of the heart, liver, and kidneys. In this case, the sudden onset and ra])idly fatal termination hidicate a true poisoning rather than an invasion of the system by bacteria, and, indeed, the poison was proved by Salkowski and Brieger to be a ptomain, to which thev irave the name mvtilotoxin. CASES ILLUSTRATIVE OF POISONING SY FISH AND MEAT, 61 Case II. — Dr. James S. Combe/ of Edinburgh, reported, in 1828, an outbreak which involved a large number of persons of the lower class ranging in age from 2 to 70 years. The first case seen was a man of 60, who complained of thirst, heat in the mouth, difficulty in swallowing, tension about the jaws and throat. The pulse was small and weak, the respiration normal, the surface cool. The hands were numb and the legs unable to support the body. Recovery followed purgative treatment. He had supped the evening before with a friend, who died during the night. They had eaten mussels boiled with salt, but had noticed no peculiarity of taste. The next case seen was that of a man of 30 who, on the previous evening, had picked a few mussels, not over five or six, and had eaten them raw. No effects were noticed until morning, excepting slight burning of the lips and tongue. On attempting to get up he found that he could not stand, although he, like the first, could move his legs about in bed. Although hundreds of cases, with many deaths, were said to have occurred, in consequence of which the magistrates issued a warning against the use of mussels. Dr. Combe found but thirty cases with two deaths. In all, the symptoms presented a striking uniformity, though they varied much in severity. Most of the victims had eaten the mussels boiled with salt and pepper, and none had noticed any unusual taste. In general the symptoms appeared in an hour or two. The man who died had vomited a few hours after eating. He lay down, had occasional general trembling, was rational to the last, and died as if by increasing weakness. On section a few dark-reel patches were found in the ileum. The stomach was empty and presented no abnormal appearance. The other fatal case was that of a woman who died in three hours after eating. The autopsy revealed a full stomach containing mussels and potatoes, and beyond a few red patches in the intestine the viscera were quite normal. In his report, Dr. Combe referred to a case related by Captain Van- couver,^ a number of whose men ate a breakfast of roasted mussels. Soon, several were seized with numbness about the face and extremities, followed by involvement of the Avhole body. One man, who died in five and a half hours after eating, was unable to swallow, and though he could row in the boat while sick, he was unable to stand on leaving it. Poisoning" by Herring's. — ^A case mvolvmg five persons, reported by R. David,^ is remarkable for the variety of manifestations, the length of time that elapsed before the appearance of the symptoms, and, in two of them, the severity and duration of the illness. The afflicted persons, adult members of one family, ate on March 19, 1898, some raw red herrings, which gave off odor indicative of commencing putre- faction. Each ate the same amount, a whole fish, but whether each fish was equally advanced in decomposition cannot, of course, be deter- mined, and the differing degrees of severity of effects may be explained ^ Edinburgh Medical and Surgical Journal, 1828, XXIX. p. 86. ^ Voyage of Discovery, Vol. IV., p. 45. ^ Deutsche medicinische Wochenschrift, 1899, No. 8. 62 FOODS. by unequal susceptibility. The father and mother aged, respectively, 65 and 67 years, suffered least; the son, aged 31, was aifected more seriously ; the two daughters presented unusually severe and compli- cated symptoms. The first effects were manifested by the sou, who, on the second day, was seized with loss of appetite, disagreeable eructations, vomiting, diarrhoea, dryness of the throat, and general weakness. On the fol- lowing day, he was better, but soon became worse. Diarrhoea was followed by obstinate constipation, which finally yielded to cathartics. Five days later, he had dimness of sight, which was followed after a week by doubk' vision and difficult deglutition. The symptoms gradu- ally abated, and on May 2 7th tiiere was distinct improvement of sight. On June 2d glasses were hardly needed. The mother first showed symptoms on the fifth day, when nausea, constipation, and dryness of the throat appeared. Several days later she had double vision and difficult deglutition. The father's case began on the ninth day and presented similar symp- toms, which disappeared in six weeks. One of the daughters was seized on the third day with bad taste in the mouth, constipation, and dryness of the throat, followed in six days by dimness of near vision, then by double vision, paralysis of accom- modation, and difficult swallowing. As was the case with the others, the temperature, circulation, and urine remained normal. On INlay 2d^ there was complete inability to swallow and it was necessary to intro- duce food by means of a stomach-tube. There was slight ptosis of the right eye, then of both ; the voice was nasal ; the gait was affected and the pulse became veiy small, though not very rapid. On May 9th, bladder symptoms, which had been gradually ai)pearing, cul- minated in paralysis of that org-an, and after the 13th, a variety of bladder and alxlominal symj)toms appeared. In the first part of July, she felt completely well, but a month later she suffered a slight relapse, with reiipjiea ranee of constipation, difficult deglutition, and disturbance of vision, which persisted with varying intensity into Sep- tember. Com])lete recovery did not occur until October, almost seven months after the initial sym[)toms. The other daughter first showed symptoms after the lapse of a week. These were in the main like those of her sister, but were more severe and extensive. She began to improve in INIav, and then ensued alternate im])rovement and loss of ground, better one day and worse the next. On the loth, there was pain in the left hypochondrium ; cm the 17th, an eruption like that of scarlet fever over the whole body, with albuminuria, but no casts. On the 1 9th, severe pain in the left hypochondrium, less in the right, and tenderness in the region of the kidneys, with epistaxis, disapjiearance of the rash, slight desquamation, and improved vision. At the end of ]May, the albuminuria and pain in the region of the kidneys had nearly disa|)peared, and deglutition was perfect. On June 2d, heart complications appeared, which persisted into November, when hypertrophy was established. In CASES ILLUSTRATIVE OF POISOXISG BY FISH ASD MEAT. 63 Auo;ust, after a general improvement, there was a relapse like that which occurred in the case of her sister. Improvement was well established in October, and in November she had almost wholly recovered. Unfortunately, it was impossible to make a bacteriological and chem- ical examination of the fish, because no material was obtainable. Poisoning by Salmon. — Professor Vaughan^ reports the following case : " K., a very vigorous man of 34 years, ate freely of canned salmon. Others at the table mth him remarked that the taste of the salmon was peculiar, and refrained from eating it. Tweh"e hours later, K. began to suifer from nausea, vomitmg, and a griping pain in the abdomen. Eighteen hours after he had eaten the fish, the writer saw liim. He was vomiting small cjuautities of mucus, colored with bile, at frequent intervals. The bowels had not moved and the griping pain continued. He was covered with a scarlatinous rash from head to foot. His pulse was 140, temperature 102° F., and respiration shal- low and irregular." After appropriate treatment he began to improve. " The next day the rash disappeared, but the temperature remained above the normal for four or five days, and it was not until a week later that the man was able to leave his house." Vaughan examiued the salmon and found a micrococcus present in great numbers. This organism, grown for twenty days ui a sterilized egg, produced a most potent poison. The white became thin, watery, and markedly alkaline, and ten drops sufficed to kill white rats. Poisoning by Oysters. — Case I. — The following case, which ended fatally, is reported by Brosch.' An officer ate a ntimber of oysters toward midnight, and within (j hours was seized with headache, paiu in the side, nausea, dimness of sight, difficult deglutition, retention of urine, and salivation. Toward noon, right facial paralysis, dilatation of the right pttpil, and thickness of speech appeared, followed shortly by cyanosis, ptosis of the right eyelid, great muscidar relaxation, and paralysis of respu'ation. Autopsy revealed pimctiform ecchymoses in various parts, enlargement of the spleen, and fatty degeneration of the liver ancT kidneys. Case II. — Another fatal case is recorded by Casey :^ ''H. P., about 32 years of age, ate 8 oysters for supper, remarking at the time that one of them was bad. Others of the same lot appeared to be C£uite fresh and were eaten by other persons with impunits'. S\Tnptoms of ])oisoning began about 12—14 hotirs later, with pam in the back, soon followed by violent pains in the stomach, freqtient vomiting, and intense thirst. The bowels did not act. These symptoms continued until the following morning, when the ptdse, which had been small and quick, became almost imperceptible, the fijigers shrunken, the nails blue. The tongue was at that time dark and swollen, and swallowmg difficult. There were occasional spasms of the arms. A little later, the jaw ^ Ptomains, Leucomains, Toxins, and Antitoxins, 1896, p. 56. ^ Wiener klinLsche AVochenschrift, 1S96, Xo. 13. 3 British Medical Journal, March 3, 1894, p. 463. 64 FOODS. became set, and soon, after a sudden stniggle for breath, he died, 41 hours after eating the oysters. At the post-ruortem examination, the heart was found to be very soft and relaxed and contained fluid blood. The kidneys and spleen were also veiy soft and congested ; the stomach emptv and darklv congested ; the peritoneum was thickly studded with flecks (.f lymph."" Poisoning by Veal. — Boyer ^ reports the following case of sextuple poisoning by \"eal. The persons afiected were members of one household, and ranged widely in pomt of age, the yoimgest being children of 3 and 6 years. The symptoms appeared in the night, about 6 hours after the food was taken, and began with vomiting and violent ailic. In the morning, all had intense gastric irritability, coated tongue, pain on pressure, especially in the right iliac fossa, rumbling, slight tympanites, and scanty tirine. The cook had markedly dilated pupils, a sensation of suflbcation, constriction and diyness of the throat, and intense suffusion of the face. The child of (] had dilated pupils and disturbance of vision, and Anally pain and stiffness of the muscles of the neck. The younger of the two children and the mother were affected less than the others, and made a more rapid recovery. The chambermaid had at first a certain degree of aggravation of syni])- toms, with a tendency to syncoj)e and great muscular weakness, which latter effects were marked also in the case of the cook, who continued for some time to be troubled by dilatation of the pupils and disturbed vision. At the end of nine days, there was no evidence of danger, and the two most severely aft'ected were well on the way to recovery. Unfortiniately, no bacteriological examination Avas made either of the meat or the discharges, but the nature of the symptoms leaves no room for doubt as to their cause. Case II. — Drs. AMlkinsou,'- Ashton, and Durham have recorded an extensive outbreak of poisoning due to imperfectly cooked veal pies. All the Citses, over fifty in number, presented very similar symp- toms, the chief of which were severe and imcontrollable vomiting and diarrhoea, accompanied at first by shivering, and followed by collapse. In some there were violent abdominal pains, and in several the abdo- men was swollen and tender. Many had severe pains in the back. The symptoms begim in from 5 to 14 hours after eating, and, as a rule, were severe from the start. The motions were first grass-green, then dark green, and highly offensive. The severity of the diarrhoea in- creased on the second day ; one patient was purged 40 or more times in a single day. In very few cases, the dejecta contained a little blood. In the worst cases, the patients became semi-comatose, restless, and delirious in the course of a few hours. Occasionally, there were dis- turl)an('es of vision, which lasted until the temjierature, which ranged from 100° in the mildest to 104.")° F. in the severest cases, became nor- mal. The pulse was very rapid, weak, and dicrotic. Many of the pa- ' Lvon im'flical, Mav 14, 1809. = Piiblic Health, January, 1899, and British Medical Journal, December 17, 1898. CASES ILLUSTRATIVE OF POISONING BY FISH AND MEAT. 65 tients were markedly cyanotic and had more or less difficulty in breath- ing. Some had cramps, and nearly all had muscular pain and stiffness. In very many cases, herpes appeared about the lips on the third to the sixth day, and some had a rash followed by desquamation. Con- valescence in the severe cases was prolonged ; some were still weak after three and a half months. Four cases terminated fatally, and in two of these, autopsies were secured. The brain surface showed slight congestion ; the small intestines showed congested patches, which be- came larger and more numerous lower down, and did not correspond with Peyer's patches. The whole lower third was highly congested, and contained yellow diarrhoeic fluid. Otherwise the organs of the body were in a fairly healthy condition. Investigation of the cause of the outbreak yielded the following facts : On July 26th, an apparently healthy calf was slaughtered, and two days later the fore quarter and breast were delivered to a baker, who made the meat into the pies which were shown to have been the cause of the outbreak. Other portions of the animal were sold to others, who made pies which caused no trouble. A portion of a knuckle •end, which was in the possession of the butcher when the investigation was begun, was to all appearances perfectly good. The baker to whom the trouble was traced made, on the day he received the meat, 160 veal pies and 108 pork pies. The pastry was the same for the entire lot, and both kinds were treated to the same lot of jelly, which was made by boiling the veal bones with two pigs' - feet in 4 quarts of water. Inasmuch as the pork pies caused no disturbance of any kind, no responsibility could be attached to the pastry or to the jelly. The veal pies were baked in not less than 3 nor more than 5 batches, hence the batches would have included about 32, 42, or 53 pies. The time occupied in baking each batch was said to have been about 20 minutes. The number of persons affected was over 50 and as in some cases single pies were shared by 2, 3, and 4 persons, it is obvious that less than 50 pies caused all the trouble. Since no other parts of the animal caused any sickness, there can be no doubt that the contamination of the meat occurred after the sale and delivery. According to the findings of Dr. Durham, based on a study of the blood of a number of the patients as to the behavior of the serum when tested for clumping properties with various micro-organisms, with controls of serum from normal persons, the outbreak was due to B. enteriticUs. This limitation of the inquiry was necessitated by the fact that it Avas impossible to secure either one of the pies, or part of one, or any of the first vomitings. The conclusion arrived at, strengthened by the fact that all 4 fatal cases were from pies which were 2 or more days old when eaten, which period allowed enormous multiplication, makes most probable the further conclusion that one whole batch was cooked so insufficiently as to preclude the killing of the organisms, which, according to Basenau, cannot survive exposure for 1 minute to a temperature of 70° C. 66 FOODS. Poisoning- by Pork. — Case I. — Meredith Youno- ' records a case of p()rk-])oisoniiio: in which 5 persons were affected. The offending meat was three-quarters of a pound of " pig's cheek," which was eaten at half-past four in the afternoon, between which time and the onset of symptoms nothing else was eaten. On the following morning, ]Mr. A. was seized suddenly with vomiting, purging, and severe abdominal pain, and shortly afterward became very feverish and weak, and suffered from severe frontal headache. His Avife had severe abdominal pain, and toward noon was strongly purged. She suffered nausea, retched, but could not vomit, had fever and severe headache, and was much more prostrated and took more time to recover than her husband. She was imable to ingest food for 8 days. The daughter was taken sick at the same time and with the same symptoms, though less severely. Her chief symptom was an overpowering tendency to sleep. A fourth person, who ate but little as compared with the amounts ingested by the others, was purged slightly, but suffering nothing more. The remainiu"; member showed no effects until durinii- the second nioht. On the following morning, she was feverish, had severe headache and abdominal pain, and retched unsuccessfully. Purging did not occur until the afternoon. As was the case with the danghter, the most prominent syinj)tom after the onset was somnolence. Kecovery fol- lowed in every case. Investigation showed that the cheeks had been cooked 2 days before, and had been placed together to cool and " set." It was estimated that between 50 and 60 persons had purchased of them, but all bnt a small proportion were unknown to the seller, and so no systematic in(juiry could be made. Only 4 could be foHowed up, and 2 of these reported no trouble ; a third was made severely sick and lost 2 days' work, and the fourth, after eating, drank so much beer that he was made sick and lost it all by vomiting, and yet was affected like the others, but not so actively. It was impossible to procure any of the meat or vomited matter or dejections for bacterio- logical examination. Case II. — At the Seventh International Medical Congress, held in London, in 1881, Ballard" read bef(»re the section on State Medi- cine an account of a very serious outbreak, now generally known as the " Welbeck case." This involved 72 persons, who attended a sale of timber and machinery on the estate of the Duke of Portland at Wel- beck, wliich lasted from Tuesday, June 15, 1880, through the week. Eefreshments were served by the keeper of a public house, and among the articles furnished were seven hams, to which the entire trouble was traced. While many complaints were made that the ham was not sufficiently cooked, that the fat was yellowish or greenish, that it was too salt, that it " tasted queer," and that it had no true flavor of ham, many made no conqilaint, and no one said that it was tainted. Of the 72 persons seized, 4 died. The history of 3 of these follows : 1. AV. W., aged 64, ate ham on Wednesday and Friday, and ' Public Health, June, 1899. '^ Supplement to lUtli Annual Report of the Local Government Board, 1881, p. 36. CASES ILLUSTRATIVE OF POISONING BY FISH AND MEAT. 67 was seized on Friday night, when he complained of feeling cold. On Saturday morning, he ate but little and said he ached all over. In the course of the day, he suifered from vomiting and diarrhoea, with severe 'pain and cramps in the legs. The evacuations were exceedingly oifen- sive and were passed involuntarily. The pulse was 128 ; temperature not taken. On Monday, he began to collapse, and on Friday, he died. The post-mortem examination revealed little that was noteworthy, but microscopic examination of the kidneys showed parenchjinatous inflam- mation, and distention and plugging of the afferent arterioles and capil- laries of the Malpighian corpuscles by emboli of bacilli. 2. Mrs. L., aged 62, ate some scraps of the ham on Wednes- day, and was seized on Friday with faintness, diarrhoea, vomiting, and abdominal pain. On the following day she fell mto a state of collapse^ and on the following Tuesday she died. The mucous membrane of the stomach and intestines was highly congested ; otherwise the autopsy revealed nothing abnormal. 3. Mr. S., aged 37, ate four sandwiches on Thursday. In the evening he vomited, and diarrhoea began. In the morning of the following day, he complained of biumiug pain in the lower part of the' abdomen. The vomiting and purging continued. Though cold and clammy to the touch, he complainecl that he was " all on fire." He had cramps in the legs and was very restless. His mind was clear to the last. The discharges were, at first, watery and offensive, and later were dark green in color. He was very thirsty and drank freely of water. He died on the following Friday. Only a j)artial autoj)sy was made.. This revealed bright-red patches on the mucosa of the stomach. The period of incubation was accurately determined in 51 cases ; in 5 it was 12 hours or less, in 34 it was between 36 and 48, and in 4 it exceeded 48 hours. In many cases the onset was sudden, and in others it was preceded by greater or less indisposition. The most con- stant symptom was diarrhoea. " In about a third of the cases the first definite symptom was a sense of chilliness, usually with rigors or tremb- ling, in one case accompanied by dyspnoea ; in a few cases it was gid- diness with faintness, sometimes accompanied by a cold sweat and tottering ; in others the first symptom was headache or pain somewhere in the trunk of the body, e. g., in the chest, back, between the shoulders, or in the abdomen, to which part the pain, wherever it might have com- menced, subsequently extended. , " In one case the first symptom noticed was a difficulty in swallow- ing. In two cases it was intense thirst. But, however the attack may have commenced, it was usually not long before pain in the abdomen, diarrhoea, and vomiting came on, diarrhoea being of more certain occur- rence than vomiting. The pain in several cases commenced in the chest or between the shoulders, and extended first to the upper and then to the lower part of the abdomen. It was usually very severe indeed, quickly producing prostration or faintness with cold sweats. It was variously described as ' crampy,' ' burning,' ' tearing,' etc. " The cliarrhoeal discharges were in some cases quite unrestrainable, 68 FOODS. and (where a description of them coidd be obtained) were said to have been exceedingly olt'ensive, and usually of a dark color. Muscular weakness was an early and very remarkable sym2)tom in nearly all cases, and in many it was so great that the ])atient could only stand by holding on to something. Headache, sometimes severe, was a common and early symptom ; in most cases there was thirst, often intense and most distressing. The tongue, when observed, was described usually as thickly coated with a brown velvety fur, but red at the tip and edges. " In the early stage, the skin was often cold to the touch, but after- ward some fever set in, the temperature arising in some cases to 101°, 103°, and 104° F. In a few severe cases M-here the skin was actually cold, the patient complamed of heat, insisted on throwing off the bed- clothes, and was very restless. The pulse in the height of the illness became quick, counting in some cases 100 to 128. '' The above were the symptoms most frequently noted. Other symptoms occurred, however, some in a few cases, and some in only solitary cases. These I now proceed to enumerate. Excessive sweat- ing, cramps in the legs, or in both legs and arms ; convulsive flexion of the hands; acliing pain in the shoulders, joints, or extremities; a sense of stiifness of the joints ; prickling or tingling or numbness of the hands, lasting far into convalescence in some cases ; a sense of general com- pression of the skin, drowsiness, hallucinations, imperfection of vision, and intolerance of light. " In three cases (one that of a medical man) there was observed yellowness of the skin, either general or confined to the face and eyes. In one case, at a late stage of the illness, there was some jmlmonary congestion, and an attack of what was regarded as gout. In the fatal cases death Avas preceded by collapse like that of cholera, coldness of the surface, }>inched features and blueness of the fingers and toes, and around the sunken eyes. The debility of convalescence was in nearly all cases protracted to several weeks. " The mildest cases were characterized usually by little remarkable beyond the folhnving symptoms, viz., alxkmiinal pains, vomiting, diar- h(x>a, thirst, headache, and muscular weakness, any one or two of which might be absent." Investigation of the hams showed absence of trichinae and the jires- ence of a bacillus, which on inoculation into animals Avas found in most cases to produce a pneumonia. The ]>criod of incubation indicates that in these cases there was a true bacterial infection. Case III. — Another epidemic investigated by Ballard^ involved a far greater number of jicrsons and had an unusual attendant mortality, nearly 500 persons out of a population of about 100,000 (Middlesbrough) dying during the year of a peculiar form of })l('Ui-()- pneumonia. The cause of this remarkable ei)idemic was proved to be the con- sumption of what was known as "American bacon," a food product •Supplement to 18th Annual Report of the Local Government Board, 1889, p. Kill CASES ILLUSTRATIVE OF POISONING BY FISH AND MEAT. 69 pre^jarecl from imported salt pork at a number of local establishments conducted under most unsanitary conditions. Twenty samples of bacon, some obtained at shops and some at the homes of victims, were examined, and fourteen were found to be distinctly poisonous to anunals. The lesions discovered in the dead animals were of the same nature and extent of those in the organs of the persons who had died. These included destructive changes in all the principal viscera, and more particularly in the lungs. Dr. Klein discovered in the lung a short bacillus which had never before been described. Inoculation experiments on animals produced results identical with those following feeding experiments with the so-called bacon. Case IV. — A remarkable outbreak due to raw pickled ham has been recorded by Van Ermengem^ and carefully investigated by him- self and others. More than t^venty members of a musical society at Ellezelles, in Belgium, were seized with serious illness after eating the greater part of a raw pickled ham ; three died within a week, and ten lay in a critical condition. Other parts of the animal from the same pickling tub were eaten in a raw state without ill effects, and pieces of the particular ham had been consumed a short time before, also ^vithout ill effects. Only those persons who ate of the ham were seized with the very peculiar train of symptoms recorded. Most of them were seized in from 20 to 24 hours, 3 in less than that time, and a few as late as 36 hours after eating. The first symptoms were gastric pain, nausea, aud vomiting of un- digested food and gelatinous blackish matters. Instead of diarrhoea, which one would expect, there was obstinate constipation in all but 2 cases, and the first dejections, with or without cathartics, were black and viscid. In every case, in from 36 to 48 hours, there were pro- found disturbances of vision — amphodiplopia, marked dilatation of the pupils, with absence of reaction to light, ptosis of both lids, and a peculiar fixed stare. There was burning thirst with a strangling sensa- tion in the throat. SwalloAving, even of liquids, was difficult or impos- sible, and every attempt was accompanied by choking. In some instances, the saliva was suppressed and the mucous mem- brane dry and glossy. The voice was weak, and with some there was total aphonia. Dysuria and anuria were common. There was but little disturbance of respiration and circulation ; the pulse never reached over 90, respiration was quiet, temperature normal. Consciousness aud general sensibility remained unimpaired throughout, except in the fatal cases, in which alone, several hours before death, there occurred collapse, dyspnoea, small irregular pulse, light delirium, and coma. There was obstinate insomnia in many, during the first period. The extremities and trunk muscles showed neither complete paralysis nor atrophy, but there was great general muscular weakness, and slight movements caused extreme fatigue. After two or three weeks, the eye symptoms began to improve. The dilated pupils contracted, the cloudi- ness disappeared, and the half-paralyzed eyelids regained their power. ^ Zeitschrift fiir Hygiene und Infectionskrankheiten, XXVI., p. 1. 70 FOODS. Diplopia disappeared only when both eyes were fixed laterally. Par- alysis of accommodatiou lasted a long: time after the disappearance of all the other symptoms, and nonnal vision did not return until after six to eiirht months. Autopsy in two cases showed no characteristic changes in the organs, only extensive hypersemia of the kidneys, liver, and meninges, and softening and unusual friability of the stomach walls. In one, the liver showed marked degeneration, and the Ijrain punctiform hemorrhages. ^Neither the liver nor Iddnevs showed anvthins^ imusual on bacteriolooi- cal examination, but the spleen yielded an anaerobic bacillus, which proved later to be capable of causing botulism. The pig from which the ham came was killed some months pre- viously, aud what was not eaten at once was pickled in the usual way. During the time that elapsed between the pickling and the supper, the greater part of the animal had been consumed without causing any sickness, but the ham which was nearly intact was the last to be eaten, lay on the bottom of the tub, and Avas the only ])art that Avas immersed completely in the Aveak briue. AVhat Avas left of it gaA'e no odor of putridity, but had a distinct odor like that of rancid butter. That the ham had a bad taste, AAas agreed by nearly all who ate of it. It appeared normal to the eye, but Avas pale, like any meat that has been soaked some time in Avater. There was no evidence of decomposition, and no ptomains Avere detected. Bacteriological examination proved in different parts the presence of a hitherto unknown spore-bearing bacillus in great abundance, the same organism as that isolated from the spleen of one of the A'ictims. It produced an extraordinarily virulent toxin, Avhich Avas isolated by Brieger from cultures supplied l)v the discoverer, by Avhom the organ- ism Avas named Bacillus botuUnns. The toxin is rendered inert by a temperature of 60° to 70° C, therein agreeing with other bacterial toxins thus far isolated. Attempts to discoA'er the organism in the feces of A-arious animals and in filth of A'arious kinds, aud in s])ecinieus from Avhere the pig Avas raised Avere negative ui results. Feedmg-experiments, conducted on A'arious kinds of animals with the meat itj^elf and Avith aqueous triturations of it added to other foods, ])ro- duced, as a rule, fatal results Avith the same train of symptoms as above mentioned. Subcutaneous injections of the Avateiy extract produced the same resiUts as feeding-experiments. The aqueous extract kept in the dark in a sealed tube retained its properties unimpaired for 10 months, and small pieces of the meat kept in cotton-stojipered tubes Avithout S]>ecial ]>recauti(ms retained their virulence even longer. The ])oison resists the effects of putrefaction, and proved to be equally poisonous after 4 days' standing in a mixture Avith feces, decomposing blood and urine, and filtration through porcelain. A fresh filtrate, to Avhich Avere added B. pro(J>(/iofiUii, B. profeu>< liqxefaciens, B. JJuoresceni't putride.'<, and B. (■')/!, was found at the end of a Aveek to be as actiA'e as ever. Poisoning by Beef. — Case I. — In Deceniber, 1toms were abdominal pain, vomiting, ]iurgiug, great weakness, and lassitude. One man of 47 years died after hardly a day's sickness. Gunthcr examined portions of the deceased and also samples of meat and blood found in the house, and sausage and meat 1 Medical Tinu's and ( iazette, Nov. 29, 1879. ^ReviK' d'lIvLriCius 18'.i(>, p. THl. ^'Archiv furllvgiene, XX VIII., p. 146. I CASES ILLUSTRATIVE OF POISONING BY FISH AND MEAT. 75 from the shop of the butcher. From the victim's spleeu and liver he isolated B. enteritidis, but while a number of species were found in the foods, this bacterium was not detected, perhaps having perished through the influence of the other species present. Case V. — This interesting case of poisoning by sausage composed of pork and beef is related by Silberschmiclt/ and serves as an illustration of the methods commonly employed in the manufacture of sausages. Nearly fifty people were poisoned by eating a kind of sau- sage known in Switzerland as "Landjager." It is made of beef, often, also, horse meat with pig fat. The materials are chopped rather coarsely, spiced, jDut into casings, pressed flat for a day, smoked two days, dried in the air several days more and then eaten in the raw state. The sausages in this instance were made of cow beef from ani- mals that had been certified as sound by a veterinarian, and pork that had been bought about two weeks previously and kept with preserva- tive salt, and had appeared fresh and unchanged when used. In the morning of the first day that the sausages were on sale, a man and his wife ate one of them together, and both were made so sick toward evening and during the night that a physician was called. In the afternoon of the same day, 19 fishermen ate of them, and on the fol- lowing day it was reported that all of them had been made sick. In the evening, another man ate one, and it pleased hmi so much that he took one home to his wife and children. On the next day, he had abdominal pains, headache, vomiting, diarrhoea, thirst, and a chill. In the afternoon, his wife and two children who had eaten were simi- larly seized. A boatman who ate two whole sausages suffered no inconvenience beyond a little pain on the following day. Another, who was sick eighteen days and then returned to his work, was seized again ten days later with the same train of symptoms. One man, aged eighteen years, entered the hospital in the morning of the second day, and died during the night, two days and a half after ingestion of the sausage. At the tune of entrance, the abdomen was sensitive and he was passing grayish watery stools; in the afternoon, he was delirious, and his pulse was very small, irregular, and rapid. Dur- iug the night he collapsed and died. Section after twelve hours showed a spleen of normal size, swoUen mesenteric glands, and hyper- semia of the stomach and intestines. The follicles were much swollen, and in the ileum were several areas from 4 to 6 cm. in length by 1 cm. in breadth, where the mucous membrane was discolored and eroded. Other organs were normal. Six others of those affected were discharged from the hospital after from seven to fifteen days' treatment. In an adjoining town, wdiere sausages of the same lot were sold, there were 16 other cases, all with the same symptoms. Taking all the cases to- gether, the symptoms of prominence were as follows : Verv severe, partially crampy, abdominal pains; very profuse diarrhoea, the stools nmnbering from eight to twelve per day, and in color varying between gray, greenish, and yellow ; usually vomiting, the rejected matters being ^ Zeitschrift fiir Hygiene und Infectionskrankheiten, XXX., p. 328. 76 FOODS. watery and brownish ; sunken eyes, high fever, great lassitude, tender- ness over abdomen, cramps in the calves, great thirst, and, occasionally, meteorism. In most of the cases, the symptoms appeared on the day after eating. The duration of the illness ranged between one and thirty days, the greater number recovering in two weeks, and becoming fit for work in three. As is commonly the case in these outbreaks, the attention of the authorities was not drawn to the matter in either town until some days had elapsed. Chemical analyses of unused sausages were made at both places. One analyst reported negative results ; the other reported the presence of ptomains, but did not further particularize. Bacteriological investigation revealed the presence of a variety of organisms, as was to have been anticipated, and among them, especially marked, Proteus Poisoning by Kid Meat. — Hensgen ^ has reported the case of a whole family stricken after eating the meat of a kid which was killed when but a few days old. A twelve year old girl was seized in eleven hours with a chill, followed by fever, dizziness, vomiting, and violent diarrhcea. The temperature rose to 103.6° F. She was confined to her bed for five days. The father, forty-nine years old, was seized with the same symptoms in twelve to thirteen hours, and had also headache, pain in the joints, thirst, and inability to walk. The tongue was dry, the pulse rapid and small, and the pupils reacted slowly. He was sick eight days. The mother, who ate but little, was seized sud- denlv in the night with vomiting, and such great dizziness that she was unable to walk without holding on to the furniture. A boy, under two vears of age, was seized in the night with vomiting and violent diar- rhoea, which soon became bloody. The stools were unusually offensive, and jiersisted so for several days. He was sick nine days. Three other children, who ate but very little, were sick two days with slight abdominal pain and diarrhoea. No material was obtainable for exami- nation. The butcher said that the kid was apparently healthy, but the mother declared that the meat around the joints of the hind legs Avas verv soft and watery, and the joints themselves enlarged (septic poly- arthritis ?). Meat Inspection and Slaughtering. The value and advisability of thorough inspection of meats before thev are ])laced on sale are universally conceded. In this country, under the inspection law of INIarch 3, 1891, all meat intended for ex- port is required to pass a very strict system of inspection. The ani- mals are inspected before being slaughtered, and their carcasses are examined microscopically by officials of the Bureau of Animal Indus- try before being ])acked. The iusjjection of meat for local consump- tion is wholly a matter of local authority ; some States have inspct-tion laws and others have none ; many cities have special regulations \vhich are enforced bv officials who may or may not be competent through ' Zeit.scbrift fiir Fleisch- und Milchhygiene, VIII., p. 181. MEAT INSPECTION AND SLAUGHTERING. 77 proper training. In Germany, the system of inspection is very rigid, particularly in the case of meats from foreign countries. This is due very largely to the activity of the agricultural interests in protecting themselves from outside competition ; and under the benevolent plea of protecting the health of meat consmners, much care and attention are given to hunting for excuses for excluding American meats which have already been inspected. The Federal meat inspection service is, according to Salmon,^ a sanitary rather than a commercial inspection, applied not alone to meats for export, but also to those intended for inter-state commerce. Curiously, however, the very important inspection for trichinae is pri- marily a commercial matter, being applied only to pork intended for shipment to certain foreign countries which require it. The United States inspectors are instructed to condemn all female animals in an advanced stage of gestation, and to prevent their slaughter for food, Salmon ruling that, though " the animal is, strictly speaking, in a physiologic condition, it is not in its usual physiologic condition, nor is the change one which is calculated to improve the quality of the meat." Females in which parturition has recently occurred are like- wise condemned as unfit for food. Many animals are condemned on account of bruises and injuries received on their way to market ; during 1900, there were condemned for this cause, in round numbers, carcasses or parts of carcasses of 4500 cattle, 1,000 sheep, and 12,300 hogs. In some of these, the injuries were extensive, sometimes complicated with abscesses, septic infection, and gangrene. The cattle diseases most prominent as causes of condemnation are tuberculosis, actinomycosis, and anaemia ; next in order are septicsemia, pneumonia, peritonitis, pysemia, icterus, abscesses, and Texas fever. In swine, the most common diseases are hog cholera, s\\dne plague, tubercu- losis, icterus, pyeemia, abscesses, pneumonia, inflammations of the abdominal cavity, septicsemia, and tumors. The most common causes of condemnation of sheep are anaemia and emaciation, bruises and injuries, tuberculosis, abscesses, pneumonia, uraemia, septicaemia, icterus, and pyaemia. In by no means every case is the entire carcass of an animal afflicted with tuberculosis or actinomycosis condemned, since, in the early stages, both diseases usually are localized, and the carcass as a whole not affected. A tuberculous animal is condemned wholly when there is emaciation or generalization of the lesions, and " when the lesions in any organ or organs are of such number and size as to indicate that the system at large may have been affected, either by inflammation, by the mixed infection, by the secretion and absorption of pus or toxic principles, or by interference with the general nutrition of the body " (Salmon). In nine years of Federal meat inspection, the condemna- tions per 10,000 animals, were, according to Salmon,^ as follows : cattle, 0.48 ; sheep, 8.1 ; swine, 37. ^ Journal of the American Medical Association, Dec. 28, 1901, p. 1715. ^ Ibidem, March 30, 1901. 78 FOODS. In inspecting meats, special attention shonld be paid to the connective tissue and glandular organs. The odor of a carcass should be sweet, and the meat should communicate no unpleasant smell to a wooden skewer thrust into it and withdrawn. The muscle should be firm and elastic, but not tough. Any variation from the natural color should be regarded Avith suspicion, verv^ dark color suggesting febrile condition, or that the animal was not slaughtered, or Avas slaughtered in a dying condition. Such meat undergoes decomposition much more raj)idly than normal meat. Animals that have been drowned or have been killed by accident without being bled yield a dark and discolored meat that is likely to decomjiose more ra])idly than that of animals that have regularly been slaughtered, but an animal that has been injured, but not killed, may be slaughtered, ])roperly l)led and dressed, and its mejit is then perfectly good. Animals should be kept without food for at least tA\elve hour^< before sltuigbter, and the carcasses should be hung for a number (»f hours to cool. Many diseases are indicated more clearly after the body has cooled. The Jewish method of slaughtering is regarded by many as far superior to any other. According to Dembo/ it is the most rational from a hygienic stand])oint, since tlie animal is l)led ra])idly and com- pletely, and the convulsive movements cause the meat to be more tender and of more attractive appearance. Lactic acid is developed, and through its chemical action on potassium ])hosphate, potassium lactate and acid ]ihosj)hate of potassium are foniied. The latter hinders the development of micro-organisms, delays the formation of ptomains and other poisonous matters, and improves the taste. Rigor mortis comes on more quickly, and the meat is, therefore, more quickly available for use, and also will keep several days longer than ordinarily. A process of slaughtering originating in Denmark appears to have borne the test of a hard three-months' trial in a very satisfactory manner, and recommends itself for adoption in the tropics, where meats decom]M)se Avith exceeding rapidity. The animal is shot in the forehead and killed or stunned, and as it falls, an incision is made over the heart and the ventricle is opened for two purposes : to allow the blood to escape, and to admit of the injection of a solution of salt through the bloodvessels by the aid of a powerful syringe. The process requires but a few minutes, and the carcass may be cut up at ouce. EGGS. Eggs form a valuable substitute for meats, being fairly rich in fats and proteids, and are well ada])ted to the stomach of the invalid and convalescent Avhen meats cannot be borne. The nutritive ])art of the white is practically limited to proteids, which amount to about 12 per cent. ; the yolk is richer in proteids, and contains in addition about 33 'Deutsche Vierteljahi-schiift fiir offentliehe Gesnndheitspflege, XXVI., p. 688. EGGS. 79 per cent, of fat. The albumin of the white is in a condition of solu- tion in cells with very thin walls. The fatty matters of the yolk are in a condition of emulsion, being held in suspension by the vitelhn. The entire yolk is held together by an enveloping membrane and is sus- pended in the white, being held in position by an albuminous band at either end : The following table by Langworthy ^ shows the average composition of eggs of different sorts : Fuel Refuse. Water. Protein. Fat. Ash. value per pound. Hen: Per cent. Per cent. Per cent. Per cent. Per cent. Calories. Whole egg as purchased .... ■ 11.2 65.5 11.9 9.3 0.9 635 Whole egg, edible portion . . . 73.7 13.4 10.5 1.0 720 White 86.2 49.5 73.3 12.3 15.7 13.2 2 33; 3 12.0 .6 1.1 .8 250 Yolk 1,705 Whole egg boiled, edible portion 765 White-shelled eggs as purchased '16.7 65.6 11.8 10.8 .6 675 Brown-shelled eggs as purchased 10.9 64.8 11.9 11.2 .7 695 Duck: Whole egg as purchased .... 13.7 60.8 12.1 12.5 .8 750 Whole egg, edible portion . . 70.5 13.3 14.5 1.0 860 White 87.0 11.1 .03 .8 210 Yolk 45.8 16.8 36.2 1.2 1,840 Goose : Whole egg as purchased .... 14.2 59.7 12.9 12.3 .9 760 Whole egg, edible portion . . . 69.5 13.8 14.4 1.0 865 White 86.3 44.1 63.5 11.6 17.3 12 2 .02 36.2 9.7 .8 1.3 .8 215 Yolk 1,850 Turkey : Whole egg as purchased .... 13.8 635 Whole egg, edible portion . . . 73.7 13;4 11.2 .9 720 ■ White 86.7 48.3 11.5 17.4 .03 32.9 .8 1.2 215 Yolk 1,710 Guinea fowl : Whole egg as purchased .... 16.9 60.5 11.9 9.9 .8 640 Whole egg, edible portion . . . 72.8 13.5 12.0 .9 755 White 86.6 49.7 11.6 16.7 .03 31.8 .8 1.2 215 Yolk 1,655 Plover : Whole egg as purchased .... 9.6 67.3 9.7 10.6 .9 625 Whole egg, edible portion . . . 74.4 10.7 11.7 1.0 695 Evaporated hens' eggs 6.4 46.9 36.0 3.6 2525 The proteids of eggs have been studied by Osborne and Campbell,^ who found that the yolk contains a large amount of protein which resembles a globulin, but is believed to be a mixture of compounds of protein matter with lecithin. The proteids of the white were found to include ovalbumin, ovomucin, ovomucoid, and conalbumin. Eggs contain a certain amount of sulphur, to which the staining of silver spoons and the odor of rotten eggs (hydrogen sulphide) are due. The rotting of eggs is supposed to be due to the admission of fermenta- tive micro-organisms through the pores of the shell, or to those already present before the shell is formed. It is a commonly accepted idea in some parts of the country that eggs with brown shells are of greater richness than others, and that the degree of richness is directly proportionate to the depth of color. In some markets, on the other hand, the white egg is held in higher esteem. ^U. S. Department of Agriculture, Farmers' Bulletin, No. 128 (1901). ^ Report of Connecticut Experiment Station, 1899, p. 339. 80 FOODS. According to the results of an extensive study of the chemical composi- tion of eggs carried on at the California Experiment Station mainly for the purpose of determining what diiferences, if any, exist between them, there is no basis of fact for the popular belief. In fact, the very slight diflFerences noted were in favor of the white eggs, but the average dif- ferences between the two kinds were less than the fluctuations between individual specimens of the same group. The figures obtained are presented in the following table taken from Farmers' Bulletin No. 87 :' ANALYSIS OF BROWN-SHELLED AND WHITE-SHELLED EGGS. Brovm-shelled eggs: Yolk White 86.60 Entire egg 65.57 Yolk White . . Entire egg While-shelled eggs: Water. Protein. Fat. Ash. SheU. Total. Per ct. Per ct. Per ct. Per ct. Per ct. Per ct. 49.59 15.58 33.52 1.04 99.73 86.60 11.99 .21 .54 99.34 65.57 11.84 10.77 .64 10.70 99.52 49.81 15.49 33.34 1.05 99.69 86.37 12.14 .35 .56 . . 1 99.42 64.79 11.92 11.22 .67 10.92 99.52 The question of influence of breed on composition has been investi- gated at the Michigan Experiment Station. The results showed that the variations in composition are too slight to be of practical value, and, as with the brown and the M'hite eggs, so slight as to be less than the variations between individual specimens from the same breed. The influence of the nature of the feed was investigated also, and was found to be of little or no importance. The flavor of eggs varies according to age, those which are per- fectly fresh having the finest flavor. It is dependent also, to some extent, upon the nature of the food consumed by the fowl, the best coming from a purely grain feed. A ver}' nitrogenous feed causes a more or less disagreeable flavor and odor. The influence of highly flavored feed has been studied by Emery,^ who fed hens with a ration containing wild onion tops and bulbs. After fifteen days, the eggs having no unusual taste, each hen received daily one ounce of this addition instead of a half ounce as before, and in three days the eggs were flavored so strongly as to be repugnant to the taste. The iron content of the yolk of eggs is said by Schmidt^ to be in- creased materially by feeding saccharate of iron to hens. He asserts, also, that the iron so incorporated is more assimilable than most iron preparations given in the aniemic condition. Aufsberg * asserts that by feeding certain iron compounds, the iron content can be increased eight times. The digestibility of eggs has been studied at the Minnesota Experi- ^ Government Printing Office, Washington, 1899, p. 24. 'Bulletin 167, North Carolina Experiment Station. * Zeitschrift fiir angewandte Cheniie, 1900, p. 705. * Pharmaceutisohe Zeitiing, 1900, p. 366. LARD. 81 ment Station.^ It was shown that, while the method of cooking has some effect on the rate of digestion, the total digestibility is not affected. Eggs boiled three, five, and twenty minutes, and digested for five hours with pepsin solution, showed at the expiration of that time respectively 8.3, 3.9, and 4.1 per cent, of undigested proteids. Cooked for five and ten minutes in water at 180° F. and similarly treated, they left no un- digested residuum. LARD. Lard is the semi-solid fat of the slaughtered hog, separated from the tissues by the aid of heat. According to the parts from which it is de- rived, it is classified as follows : (1) Neutral lard. This is derived from the fresh leaf, which is reduced to a pulp after being cooled, and then rendered in the kettle. A part of the fat is separated at from 105° to 120° F., and the residue is sent to the rendering tanks for further treatment. The lard obtained is washed, while hot, with water containing a trace of sodium carbonate, common salt, or dilute acid. (2) Leaf lard. This is obtained from the residue above mentioned, which is subjected to steam heat under pressure. (3) Choice kettle rendered lard. This is obtained from the remaining portions of the leaf together with the fat from the backs. Both the leaf and back fat are passed first through a pulping machine. (4) Prime steam lard. This is made from the head, the fat of the small intestine, trimmings, and other fatty parts. The spleen, pancreas, trachea, and all other refuse parts and trim- mings, with the exception of the small intestine, the liver, lungs, and part of the heart, go into the rendering-kettle for what fat there may be in them, and the product is variously, but not graphically, desig- nated. " Refined lard " is a term used to designate a lard composed chiefly of cotton oil and stearin. It is known more often as "lard com- pound." Physical and Chemical Properties of Lard. — At 40° F., the specific gravity is 0.890; at 100°, about 0.860 ; it differs not very materially from that of the substances used as adulterants, excepting cotton-seed oil, which is notablv heavier. The melting-point ranges from 39.1° to 44.9° C. (102.4°'to 112.8° F.), according to the part of the carcass from which the fat is derived, and hence it cannot be taken as a safe guide in the determination of purity. Pure lard, melted, and mixed with strong sulphuric or nitric acid, will give only a slight color, which may be yellowish, pinkish, or inclined to light bro^vmish. Cotton-seed oil and other seed oils, and mixtures containing them, similarly treated, yield any color between yellowish brown and very brownish black or even black. The re- fractive index of pure lard is materially lower than that of cotton- seed oil. Pure lard contains only traces of volatile fatty acids, 5 grams yield- ^ Farmers' Bulletin No. 87, Government Printing Office, Washington, 1899, p. 25. 82 FOODS. mg an amount which i;? neutralized by 4 or | of a ce. of deeinormal sodium hydrate sohition. The non-volatile fatty acids are present to the extent of about 95 }3er cent. The iodine absorption number varies according to the part of the carcass from Avhich the fat is derived, but averages about 60. The iodine number of cotton-seed oil is about 109, and that of stearin is approximately 20. Thus, these substances used as adulterants may be mixed in such proportion as to yield the normal iodine number of lard. With nitrate of silver solution, pure lard causes no more than the very slightest amount of reduction, and generally none at all ; but cotton- seed oil causes a very marked reduction of the salt to the metallic state, with the result that the mixture has a brownish or black appearance from the minute black particles formed. A small amount of lard, dissolved in a mixture of equal parts of alcohol and strong ether in a test-tube and allowed to stand in a cool place, will, when the solvent in large part is evaporated, show masses of crystals, which, on examination under the microscope, are seen to be rhombic and extremely variable iu size. Beef stearin, similarly treated, shows fan-shaped and dumbbell-shaped clusters of needle crystals. INIixtures of pure lard and beef stearin will show both forms of crys- tals. Sometimes, when ciystallization proceeds rapidly, the ciystals from pure lard are extremely small, and are clustered in such a way as to be distinguished from beef stearin crystals only with great difficulty. It is essential that the ciystallizing process shall proceed slowly, and that the amount of lard dissolved in half a test-tube of the solvent shall be quite small — not larger than a large pea. The mouth of the test-tube should be stopped with cotton. Section 3. MILK AND MILK PRODUCTS. MILK. Milk is a solution of sugar, mineral matter, and proteids, with other proteids and fat in suspension. Its composition is very variable, not alone as between different species of mammalia by which it is produced, but as between different individuals of the same species. Of the domestic animals, the ass and mare produce milk which most closely approximates that of woman in composition, but our chief interest in milk as an article of food in general use lies in that produced by cows and, to a certain extent, in that of goats, which is very similar m com- position. AVliile the composition of milk of other auimals than those already mentioned can have for most of us merely a scientific interest, it may be of some practical utility in the management of breast-milk to bear in mind that the milk of animals whose diet is largely or chiefly meat is richest in those elements, the proteids, that are most conmiouly at the bottom of digestive disturbances in breast-fed children. Composition of Cows' Milk. — The composition of milk of average good quality may be expressed fairly in round numbers as follows : MILK. 83; Fat 4.00 Sugar 5.00 Proteids 3.30 Mineral matter 0.7 Total solids 13.00 Water 87.00 100.00 According to Vieth, the average composition of more than 120,000 samples analyzed in England was : Fat 4.10 Solids not fat 8.80 Total solids 12.90 "Water 87.10 100.00 The average of a large number of analyses made in this country showed : ^ Fat 4.00 Sugar 4.95 Proteids 3.30 Mineral matter 0.75 Total solids 13.00 Water 87.00 100.00 The milk yielded by 426 cows from private farms in Massachusetts^ and by 1 75 more belonging to public institutions, was analyzed by the author and his associates, and found to give the following results : ^ 426 cows from private farms, total solids 13.36 175 cows from public institutions, total solids 13.00 601 cows (both classes), total solids 13.26 Fat. — The fat of milk exists in very minute globules which vary widely in size, the largest being between six and seven times larger than the smallest, but the latter are most abundant. Whether or not they have an albuminous envelope, is a matter of doubt, the evidence for and against bemg about equal, and of no great importance. It consists of glycerides of ten different fatty acids, five of which belong to the non-volatile and five to the volatile class. The glycerides of the former group constitute by far the greater part. They are stearin, palmitin, olein, myristin, and butin ; the two last are present in very minute amounts. Those of the latter group give the character- istic butter flavor. They are butyrin, caproin, caprylin, caprin, and laurin ; the two first are the important ones, and together amount to over 7 per cent, of the whole fat ; the three others are present in but insignificant traces. ^ American Experiment Station Kecord, V., No. 10. ^ The detailed analyses, with data as to breed, nature, and amount of feed, etc., can be found in the pamphlet issued by the State Board of Health : Results of Inquii-ies Relative to the Quality of Milk as Produced in Massachusetts. Boston : February, 1887. 84 FOODS. Tbe fat, being the lightest part of milk, tendsj to rise to the surface when the milk is allowed to stand, and then forms a layer which we know as cream. This contains not fat alone, but all of the constituents of the milk, and is, therefore, simply milk containing an excessive amount of fat. It is a common error to regard the depth of the cream layer which forms on standing a given length of time as an infallible measure of the richness of the milk by which it is yielded ; but cream does not always rise well in rich milk, even after standing more than twenty- four hours. The author repeatedly has found the percentage of cream thrown up by a specimen of milk in a 100 cc. graduate in twenty-four hours, as measured by the lines of graduation, to be less than the actual percentage of fat as shown by analysis. The rapidity with which the fat finds its way to the surface depends largely upon the size of the fat globules. The largest rise first, and the very smallest may not rise at all. Again, a watered milk throws up its fat more quickly than a normal specimen, although it does not contain as much. It appears, therefore, that a millv of inferior grade may under some circumstances show a deeper cream layer than a milk of unusual richness. Generally speaking, however, a rich milk will usually show its quality on standing. The first part of a milking is always poor in fat, the middle portion contains about the average amount of the whole, and the last portion is always the richest. The first portion is known as " fore-milk," the last as " strippings." A specimen of '' strippings/' analyzed by the author, gave the following results : Fat 9.82 Sugar 4.00 Proteids 4.21 Asli 0.79 Total solids 18.82 Water 81.18 100.00 Milk-sugar. — Lactose or sugar of milk, is peculiar to milk. It is much less soluble in Avater than dextrose and sucrose. Heated to 100°-131° C, it becomes changed in color to brownish, and at higher temperatures loses water of crystallization and undergoes further change. At 175° C., lactocaramel is formed. When heated in solution, in milk itself, for example, it begins to undergo decom])osition changes at 70° C. and above. Through the action of the lactic ferments, it gives rise to lactic acid. In the polariscope, it is dextrorotary. Proteids. — The greater part of the proteids of milk, about 80 ]ier cent., is casein, or, as it is called sometimes, caseinogcn. It contains both sulphur and ])hosphorus, and is in intimate combination with cal- cium phosphate. It is not coagulated by heat, but is precipitated by acids, by which the combination is broken up. 1 n the presence of lactic acid in small amounts, due to the breaking up of lactose, coagu- lation is hastened by the a])plication of gentle heat. This phenouu^uon MILK. 85 is observed very commonly in the case of milk which to the taste is apparently sweet, but which is "just on the turn," The chief part of the remainder of the proteids is lactalbumin. This is coagulated by heating to 65°-73° C, but not by dilute acids. It contains sulphur, but no phosphorus. In amount it ranges from 0,2 to 0,8 per cent. It is much more abundant in colostrum. The re- maining proteids are lactoglobulin, which is coagulated by heat ; lacto- protein, coagulable by neither heat nor dilute acids, and fibrin. Each exists in but very small amounts. Mineral Matter. — The mineral matter contained in milk consists of phosphates and chlorides of potassium, sodium, calcium, and magnesium, and extremely mmute traces of iron. Of the bases, potassium is the most abundant, with calcium, sodium, and magnesium in the order given. The phosphates predominate over the chlorides. Part of the calcium exists in combination as phosphate with the casein, and the rest, according to Danilewsky,^ as mono- and tricalcium phosphate and in combination with citric acid. Part of the magnesium, also, exists ui combination with citric and other organic acids. In very small amounts, these are normal constituents of milk of various animals. In human milk, citric acid is present to the extent of about 0,05 per cent,, and in cows' milk, it is about three times as abundant. Specific Gravity, — The specific gravity of cows' milk of normal com- position ranges from 1,029 to 1.034. It increases very slightly for about five hours after the milk is drawn, and then becomes stationary. The increase is believed to be due to molecular modification of the casein, and not to the escape of gases. It is lowered by fat and water, and by the presence of bubbles of air, and is raised by removal of cream. Reaction. — When freshly drawn, milk shows the so-called amphoteric reaction ; that is, it is acid to litmus and alkaline to turmeric. The alkaline reaction is iatensified on warming, but the acid reaction is not influenced thereby. On standing, the alkaline reaction is overcome by the lactic acid which is formed gradually from the sugar, and the acid reaction is uicreased in consequence of the same. The original acid reaction is due to the presence of phosphates, the alkaline to alka- line carbonates. Human milk is normally alkaline, and that of car- nivora is acid. Appearance. — The appearance of normal milk is too familiar to need description ; but imder certain rare abnormal conditions, milk may assume different colors, including blue, yellow, violet, and red. These changes of color are due to the action of certain bacteria, and are always evidence of imsanitary conditions to which the milk is exposed at the dairy or during distribution and storage. Blue milk is due to the action of B. cyanogenes, which produces a blue color in no other food material. For its development it requires the presence of lactic ferments, and, therefore, has no eifect on milk that is sterile. Another organism capable of producing the same effect is B. cyaneofluorescens. 1 Wi-atsch, 1901, p. 549, 86 FOODS. A red color may be caused by B. prodigiosus or B. lactis erythro- genes, sometimes by blood, and, it is said, by madder and other red coloring-matter in the feed. Yellow is caused by B. synxanthus, and violet by B. violaceus. All of these abnormal milks are, aside from their uninviting appear- ance, unfit for food, since they are likely to cause gastro-intestinal irri- tation. Thus Eichert^ records a case of severe diarrhoea, with very offensive stools, in a child of nine months, which was due to red milk caused bv a bacillus (probably B. lactis erythrogenes) present in the milk ducts at the time of milking. Another abnormal condition caused by a large variety of organisms is characterized by alteration of the natural consistence to one of sliminess, which appears only some hours after the milk is drawn. Slimv or "ropy" milk throws up no cream, and cannot be churned, but altliough it is most repugnant to the senses, it causes no digestive disturbances if ingested. In three specimens of ropy milk from as many different creameries in the State of New York, Ward^ found the change due to B. lactis viscosus (Adametz). Taste. — The flavor of milk is modified very sensibly by the char- acter of the feed and by the absorption of gases and volatile matters of all kinds. It is aifected very readily by turnips, garlic, wild onion, mouldy hay and grain, distillery swill, and damaged, rotten ensilage. Bitterness of taste may be due either to some constituent of the feed or to bacteria. When due to feed, the taste is bitter from the very first, but when caused by bacterial agency it develops some time after milking, when the organisms which produce it have had the opportu- nity to act upon the proteids or whatever constituent may be concerned. It mav be due also to inflammatory conditions of the udder, in which case it may or may not be noticeable when the milk is freshly dra\\n. The bacteria concerned in producing bitterness may exist in the ducts of the teats, or may come from stable filth. Damman^ mentions a case in which the persistently bitter taste disappeared after the floor of the stiible was cleaned and disinfected, and the ducts of the teats svringed out with disinfectant solution. Strong-smelling disiiift'otants may not be used in dairies because of the readiness with which milk absorbs odors. This absorptive capacity is so well recognized that milk is stored commonly in separate compartments of refrigerators, a^\"ay from foods which evolve distinct odors. Presence of Alcohol. — Distillery swill not only causes a decicUdly bad flavor, but may, in addition, and contrary to a generally accepted idea, cause an alcoholic milk. Thus, according to H. W. Weller,^ a sample of milk derived from cows fed on distillery refuse containing 6.90 per cent, of alcohol yielded in addition to a high proportion of milk solids, 0.96 per cent, by weight of alcohol. The milk was com- ' Zeitschrift fiir Fleisch- und Milchhvgiene VIII., No. 5. 2 Science, 1001, No. 322, ]^. 324. 3 Deutsche thieriirztliclic Wochcnschrift, 1897, No. 1. * Foi-scliungsberichle iiber Jvebensinittel, etc., 1897, p. 206. MILK. 87 plained of on account of an unpleasant after-taste. Teichert^ records a case in which calves and lambs failed to thrive, and many died from a form of diarrhoea. The mothers were fed on distillery waste, and yielded milk containing alcohol. That alcohol or something connected therewith may be eliminated in milk, is shown by numerous cases, among which are the following: Vallin^ records that a nursing infant was seized mth convulsions with great regularity on Mondays and Thursdays, but was quite well on other days. In^'estigation showed that the wet-nurse on Sundays and AVednesdays, her "days out," was in the habit of drinkiug freely of alcoholics. The curtailment of the privilege was followed by disappearance of the difficulty. Farez' cites 2 cases which show the bad influence of alcohol on nursing children. In one, the wet-nurse drank wine at meals, and especially in the even- ing, and the child never ceased fretting, crs'ing, and screaming from 9 o'clock imtil 11. The nurse complained bitterly of the naughtiness of the child, and was grieved at the suggestion that she was herself at fault through drinking too much, but she was induced to abstain from alcohol entirely, and from that time there was no further trouble. In the other case, the mother drank tea at noon and ^dne at dinner, and the child was quiet during I he afternoon, but screamed and fretted all the evening and until midnight. A change to wine at noon and tea at dinner produced a corresponding change in the behavior of the child, the turbulent period occurring in the afternoon. When the mother eliminated wine from her dietary entirely, the trouble ceased. Colostrum. — The milk secreted before and in the early stage of lactation is known as colostrum. It is a yellow, somewhat viscid fluid of strong odor and acid reaction. In composition, it differs very ma- terially from milk, particularly in its percentage of j)roteids. It con- tains, sometimes, so large a percentage of lactalbmnm and lactogiobulin that it is coagulated by boiling. Its content of casein is about normal, but it is not coagulated by rennet, or at most imperfectly. In the early stages, its sugar is dextrose and not lactose. According to Tiemann,* it ranges in specific gravity from 1.0299 to 1.0594, in fat from 0.56 to 9.28, in proteids from 4.66 to 21.78, in ash from 0.82 to 1.25, and in total solids from 12.93 to 32.93. Under the microscope, it shows large corpuscles, known as colostrum corpuscles, which disappear within two weeks at most after the time of calving. Changes Produced in Milk by Boiling. — Boiling causes greater coalescence of the fat globules, changes in the character of the sugar, coagulation of lactalbumin, and destruction of micro-organisms and ferments. Boiled milk, therefore, will keep better than raw milk. The scum which forms on the surface is largely fat, casein, and lactalbu- min, and occurs in consequence of rapid evaporation at that point. Boiled milk is dio:ested sliffhtlv less readilv than raw milk, and som* 1 Milch Zeitung 1901, p. 148. ^Eevvie d'Hvgiene, 1896, p. 953. ^ Tribune m'edicale, .June 20, 1900, p. 488. *Zeitschrift fiir physiologisclie Chemie, 1898, p. 363. 88 FOODS. milk is digested more readily than either ; but boiled milk, as will be noted later (see page 92), is uot always a desirable food for yoiiug infants, on account of its changed character. The question whether or not a given milk has been boiled may readily be determined by the application of simple tests (see Analysis of Milk). Changes Due to Bacterial Action. — At ordinary temperatures, milk soon bejiins to under<>;o chanties initiated and carried alon^ bv various species of micro-organisms which exist in the ducts of the teats or fall into the pail from the external surface of the udder or surround- ing parts, or from the air, or from the hands and clothes of the milker, or which are already present m the ])ail or otlier vessel into which the milk is received. The most common change is brought about by the lactic ferments, of which more than a hundred species have already been identified. They attack the milk-sugar and cause the formation of lactic acid, which, on accmnulating in sufficient amount, causes the milk to curdle. Their multiplication proceeds most rapidly at tem- peratures ranging from 25° to 30° C Therefore, in order to inhibit their action as far as possible, milk should be cooled without delay and kept in storage at low temperature. In addition to the lactic ferments, there are others wliich are known as casein ferments. These produce a substance much like rennet in its action. They may act in the al)- sence of the lactic ferments, and then their action is accompanied by the development of alkalinity. Under certain conditions, in addition to changes in taste, color, and consistency already noted, intensely ])oisonous benzene derivatives are formed, the most important of which, diazobenzene, called by its dis- coverer. Professor V. C Vaughan, tyrotoxicon, is the exciting cause of the train of symptoms commonly known as milk poisoning, cheese poisoning, and ice-cream jioisoning. A number of other organisms constitute the group of so-called butyric ferments, many of which are of the class of casein ferments. They cause the production of butyric acid in the decomposition of j)roteids. In consequence of the action of the various species of orgsinisms, it is important that bacteria in general should l)e excluded as C()m])l('tely as ])()ssible from milk by the observance of the utmost cleanliness in milking, handling, and storing. The milk of cows stalled in badly ventilated, unclean stables, and of cows with unclean udders, will decom- pose much more rapidly than that of cows kejit under better sanitary conditions. Even Avhen the cow and her surrouudiugs are kept in a cleanly state, the very first part of a milking should be rejected, on account of the very large numbers of bacteria present in the ducts of the teats. Under even the best of conditions, many bacteria are present in freshly draAvn milk, and these increase ra])idly in number unless killed bv the action of heat or other germicides. Indeed, it has i)e('n found impossible in the majority' of experiments to obtain sterile milk even when the greatest precautions have been observed to exclude extraneous organisms. The first })art of a milking is richest in bacteria, because those which I MILK. 89 have multiplied within the ducts of the teats since the previous milldng are expelled mostly with the fore-milk, but even the very last portions of the strippmgs may contain as many as 500 bacteria per cc. Thus, Schultz^ foimd in the first portions of cows' milk 97,240 per cc, in the strippings 500, and in goats' milk 78,718 and 665. For the attain- ment of the best results as to keeping qualities, all dirt should be brushed from the cow before milking, and the udder and flank should be dampened, in order that dust, fine dirt, and bacteria may be retained in situ, and not fall into the milk-pail, which should always be perfectly clean before use. The difference in the number of bacteria which fall mto milk when proper precautions are observed and when they are neglected is very considerable. Thus, Soxhlet found that the milk of a cow with a dirty udder, stalled in a duly stable, kept sweet 50 hours at ordinary tem- perature, and that, when her udder was washed and she was milked in the open air, it remained sweet a day and a half longer. Still more instructive are the results obtained by Freeman,^ who exposed plates, 3.5 inches in diameter, for two minutes as follows : one in the open air, one inside a barn, and a third in front of the milk pail under a cow in the same barn while being milked. The first plate showed 6, the second 111, and the third 1,800 colonies. Such a number of bac- teria, falling upon so small a surface within so short a time, is an index of the enormous number which may fall into a pail during the time required for a complete milking. The enormous number of bacteria A\^hich may be commonly present in ordinary market milk, the great influence thereon of non-observance of the strictest cleanliness, and the extreme rapidity of multiplication under favoring conditions, are shown in most striking manner by W. H. Park,^ who exposes the inexcusable lack of cleanliness in the methods of procuring milk, and of care in cooling, and in keeping it dui'ing transportation to the city. Milk from individual cows, ^vhere every reasonable means was taken to insure cleanliness, yielded an aver- age of 6,000 bacteria per cc. when 5 hours old, and kept at 45° F., to which temperature it was cooled soon after it was drawn. After 24 hours, the average number fell to 1,933 ; after 48, it increased to 17,816. Milk taken in winter in well-ventilated, fairly clean, but dusty, barns, and cooled within 2 hours to 45° F., the visible dirt having been cleaned off the hair about the udder, the milkers' hands wiped off, but not washed, the pails and cans clean, but the straining cloths dusty, yielded the following average figures: At time of milking, 15,500; after 24 hours, 21,6^66; after 48 hours, 76,000. Milk taken from cows kept in ordinary barns, the conditions as to cleanliness of sur- roundings and method of milking being about what obtain on the aver- age farm, yielded the following average figures : ^ Archiv fiir Hygiene, XIV., p. 260. 2 Medical EecoVd, March 8, 1896. » Journal of Hygiene, July, 1901, p. 391. 90 FOODS. Winter. Summer. Shortly after milking 16,650 30,366 After 24 hours 31,000 48,000 After 48 houi-s 210,000 680,000 Twenty samples of average milk taken immediately on arrival in the city, much of it having been transported more than 200 miles, yielded from 52,000 to 85,200,000 bacteria per cc, (average, 5,669,- 850). The average temperature of the samples when taken from the cans was 45° F. Milk as sold in the shops during the morning hours yielded the following averages : From tenement districts, mid-winter (13 samples) .... 1,977,692 From well-torent temperatures on the rapidity of bacterial nndtiplication is avcII shown by the results obtained on allowing por- tions of the same specimen to stand under otherwise similar conditions. At temperatures below 50° F., there was at the end of 24 hours no in- crease — in fact, a decrease — in the numljcr of bacteria ; but at higher temperatures, the multiplication was enornvous. The original number per cc. was 3000, and the growths at the several temperatures above 55° F. were as follows at the end of 24 and 48 hours: Temperature. 24 hours. 48 hours. 60° F., 68° 180,000 450,000 28,000,000 25,000,000,000 86° 94° 1,400,000,000 25,000,000,000 ]\rilk of fair quality from a shop Mas kept at 90° F. for 8 hours, during which time its contained bacteria increased from 92,000 to 6,800,000 per cc. ; another, of poor qualitv, under the same conditions, showed an increase from 2,600,000 to 124,000,000. Such growths of bacteria in milk intended for human use can iu no way improve the milk, but must seriously affi^ct its wholesomeness. To avoid them, the means are simple : cleanliness everywhere and low temperatures; cleanliness of the cows' exterior, of the stable, of the milkers and their clothing, of all vessels employed — milk pails, pans, bottles, etc. — and of the places where the milk is stored. MILK. 91 Preservation of Milk. — The keeping quality of milk is influenced by cold, ^vhich retards the growth and multiplication of bacteria which bring about decomposition ; by heat, which destroys them ; and by pre- servatives, which either kill them or retard their gro^^*th. Preservation by cold is in many respects preferable to either of the other methods. The constituents are in no way altered in character, there is no change in digestibility, and no element is introduced into the system with the milk to exert any harmful influence upon the digestive processes. In places where ice is expensive or not obtainable, this method is not available, but where it is cheap and plentiful, it is the one in most com- mon use. In some parts of Europe, milk is frozen into solid blocks by the ammonia process, and shipped in that form to market. A large part of the milk supply of Copenhagen is received from a distance in large air-tight cans, into each of which a block of frozen milk, weighing about 25 pounds, is placed, to keep the milk in which it floats at a low temperature. Presers'ation by heat includes pasteurization and sterilization. In pasteurization, the whole bulk of the milk is heated to not over 158° F., maintained at that temperatiu'e for 10 or 15 minutes, and then cooled rapidly in order to preserve the fresh flavor and prevent the mtiltiplication of stich of the bacteria as sur%T.ve. The length of time requhed for the destruction of bacteria varies T\ith the temperature employed. Thus, about 70 per cent, of saproph}i;ic bacteria are killed in an hour at 140° F., in 15 minutes at 150°, in 10 minutes at 158°, in 5 minutes at 176°, in 2 minutes at 194°, and in 1 minute at 203°. It is essential that all apparatus and vessels used in cooling and storing shall be clean and sterile. This process is quite sufficient for all prac- tical purposes and hygienic requirements, unless the milk is to be kept for a longer time than usual, in which case it should be repeated at the end of 24 hours. Temperatures higher than 158° F. cause the milk to acquire a cooked flavor, which to many persons is disagree- able. Accorchng to H. Bitter,^ all pathogenic bacteria in milk are killed with absolute certainty by exposure to 154.4° F. for a half hour, and the milk is altered thereby in neither appearance nor taste. Under ordinar}' circmri stances 20 minutes' exposure is Cjuite sufficient. Some authorities assert that temperatures of 140° to 147° F. are sufficiently high for the ptirpose, but Professor Theobald Smith ^ has sho^^m that, while tubercle bacilli are destroyed within 20 minutes at 140° F., the formation of a surface pellicle into which they are carried by fat globules shields them from the heat, so that they may survive an exposure of over an hour to 149° F. It is asserted by Morgenroth-^ that at least 30 mintites' exposure to 158° F. is necessary to kill all of the bacilli, but that the same result can be attained at a much lower temperature (131° F.) in 3 hours' heating in a thermophore. It has 1 Zeitschrift fiir Hyo-iene, YIII., p. 240. ^ Journal of Experimental Medicine, 1899, p. 217. ^ Hygienische Kundschau, Sept. 15, 1900, p. 865. 92 FOODS. been asserted also by M. Beck^ that 158° F., and even 176° F., are not sufficiently high, even when maintained 30 minutes, for the destruc- tion of all the tubercle bacilli in milk intentionally infected. He heated such milk for 30 minutes at both of the above temperatures, and then injected it into 15 guinea-pigs, all of which became tubercu- lous after 5 to 8 weeks. But Levy and Bruns,- after experimenting with milk enclosed in flasks placed in a water-bath, found that, so far as the tubercle bacillus is concerned, milk is sterilized after 15 to 25 minutes' exjiosure to 149°— 158° F. It seems probable, on the whole, that the widely divergent results of various experimenters have been due to diiferences in manipulation and in physical conditions. All the lactic ferments are destroyed very easily, but some of the casein femients are veiy resistant, and their spores still more so, and are not killed by boiling for a number of hours ; and it is to the pres- ence of these hardy varieties that the difficulty of complete sterilization is due. Sterilization requires continuous heating under pressure for about two hours at 248° F., at which temperature not alone the bacteria and their spores are destroyed, but the normal appearance and taste of the milk as well. Part of the sug-ar is converted to caramel, part of the casein is precipitated, and the milk will no longer form a cohesive coagulum with rennet. In the opinion of many practitioners, neither pasteurization nor sterilization is free from objection in infant feeding, since even a temper- ature of 155° F. influences the nutritive value injuriously. Many cases of sc(n'l)utus and d}'sj)e])sia in infants have been attributed to the use of sterilized milk, and it seems probalile that the trouble is con- nected %\nth the destruction of the zymases normally present. These fennents, the presence of which was announced in 1900 by Escherich, have been studied by Spolverini,'^ who isolated no less than seven, two of Avhich, pepsinic and trvjisiuic, are present always in both human and cows' milk; the others, amylolytic, lipasic, glycolytic, etc., are not constant. All are soluble, and none can withstand the sterilizing temperature. To avoid the untoward results of the use of sterilized milk. Free- man^ proposes that the cream l)e allowed to rise, and then be removed and subjected alone to sterilization, after Avhich it may be mixed in proper proportion with the skimmed milk, which contains only a very small number of bacteria, since about 99 per cent, of them are carried into the cream layer by the fat globules. Preservation of milk by the addition of antisejitics is unnecessary, unjustifiable, and possibly injurious. If millv is drawn properly from decently clean animals into clean vessels by clean milkers, and stored in clean places, it will keep sweet quite as long, under ordinary circum- ' Deutsche Vierteljahi-s-schrift fiir ofi'entliche Gesundheitspflege, XXXII. p. 430. ' Hygienisc'he Rundschau, .July 15, 1901, p. 669. ' Archive.s de Medecine des Enfant.s, Dec, 1901. * Archives of Pediatrics, August, 1899. JIILK. 93 stances and under the usual conditions of frequent delivery, as is desired by the consumer. The addition of antiseptics, which only retard growth of bacteria without destroying them, enables the vendor to supply stale milk instead of fresh, and to dispense with part of the sanitary precautions otherwise necessary. The substances used are by no means wholly innocent in their action on the human system, even in very small quantities and, moreover, it is impossible to control the amount added by a single individual or to be sure that successive handlers have not contributed additional doses. The substances used as milk preservatives are boric acid, borax, salicylic acid, formaldehyde, carbonate of sodium, and chromates. Boric acid and borax are used generally in combination with each other, experience having shown that the mixture is more efficient than either alone. The minimum efficient quantity of the mixture is about 10 grains to the quart, an amount which even for an adult may well be regarded as a fairly large medicinal dose. In addition to its action on the general system, it exerts a varying effect on the digestion according to the amount present. According to Professor R. H. Chittenden, borax retards the amylolytic action of saliva, boric acid in amounts less than 1 per cent, favors it, and both increase gastric digestion in small amounts and retard it in large. The use of salicylic acid in milk is not extensive. It is a fairly efficient preservative. Formaldehyde has come into use within a few years. It is a most efficient preservative, and not alone inhibits growth but also kills the bacteria. According to tests made by Dr. C. P. Worcester,^ 1 part of commercial formalin in 100,000 of milk will postpone the curdling-point 6 hours ; 1 in 50,000, 24 hours ; 1 in 20,- 000, 48 hours; 1 in 10,000, 138 hours; 1 in 5,000, 156 hours. Al- though nothing is known as to the action of small amounts of for- maldehyde on the general system, it is not correct to assume that, in the absence of evidence to the contrary, it is necessarily harmless or beneficial. While the occasional ingestion of a small amount of for- maldehyde may produce no effect, we cannot reason that its daily use over a long period will be equally non-productive. An occasional drink of water containing lead will do no injury, while its daily use may cause lead paralysis, and in the same way formaldehyde may be the cause of serious disturbances attributed to something else. But whether harmful or not, the use of this agent and of others is unnec- -essary and unjustifiable. Aside from its j)ossibly poisonous action, there is the objection that it alters the character of the milk proteids ; the casein becomes uncoagulable by rennet, except in thick clots, and much less digestible, or Avholly indigestible, by the proteolytic ferments. Certain it is that anything that unposes additional burdens on the digestive function of infants and invalids can hardly be regarded as a proper substance for use in food. Annet,^ after a study of formalde- ^ Twenty-ninth Annual Keport of the State Board of Health of Massachusetts, 1897, p. 559. 2 The Lancet, Nov. 11, 1899. 94 FOODS. hvde and boric acid as milk preservatives, concludes that they are in- jurious, especially to young infants, and suggests the possibility of a causal relation between their use and the great infant mortality during the hot months. Carbonate of sodium is a weak agent, and does not postpone decom- position to an extent sufficient to encourage its wide adoption. So far as is known, there can be no objection to its use on the score of injury, except in so far as the assertion that sodium lactate, fonned h\ its decomposition by the free lactic acid, acts as a mild cathartic, is worthy of credence. The chromates are not extensively employed, but have been found present in preservative powders used in France. Deniges ' found the normal chromate of p(»tassium in two of these })reparatinns, and the dichromate and chromate together in a third. The latter was recom- mended in the proportion of 2 grams to 50 hters of milk. According to Froidevaux,^ such an amount of potassium dichromate is insuffi- cient to retard coagulation and imparts an abnormal intense yellow color to the milk. The further discussion of the subject of milk preservatives may be looked for below, under the general subject of Food Preserva- tives. Adulteration of Milk. — This most important article of food is mure subject to adulteration than any other, since it lends itself so readily to fraudulent manipulation. The principal adulterations are the addition of water and the abstraction of cream. The former diminishes the nutritive value, and, if tlie water used is from an unclean source, increases the possibility of disseminating disease ; the latter robs the milk of one of its most valuable constituents. The detection of these adulterations by analysis is not always possible, since a rich milk may be slightly watered or only partially skimmed and still .show average quality. Again, even though the watering be fairly extensive, it cannot always be proved that the milk was not of low grade from natural causes, since some cows give milk Avhich on analysis is far below average good milk and bears every resemblance to watered milk. Further, a milk containing very little fat may be naturally poor in that constituent or may be the first part of a milking. In conse(iuence of the difficulty of proving the addition of water or abstraction of cream, and because of the enormous importance of secur- ing a public supply of at least average good quality, most States have fixed legal standards, to which milk intended for sale must conform. The standard for total solids is commonly 13, 12.5, or 12 percent.; iind for fat, .'>, 3.5, 3.7, and 4 per cent. By the adoption of a legal standard, all milk of low grade, whether so by reason of fraudulent practices or be use of poor feed or individual peculiarity of the cow, must be treated alike. By prohibiting the sale of all milk not of a certain grade, it becomes unneeessar}' to prove fraud or criminal knowl- ' Revue International des Falsifications, IX., p. 36. * Journal de Pliannacie et de Cheniie, 1896, p. 155. MILK. 95 edge, the allegation of inferior quality being sustained by the results of the analysis. Other forms of adulteration include the addition of coloring matters for the purpose of concealing watering or skimming, or to give a creamy tint to a very white milk, and the addition of preservatives, and, occa- sionally, of other foreign substances. The coloring matters commonly used are, annatto, caramel, and combinations of aniline dyes. Their detection is by no means difficult (see Analysis of Milk). It is a common belief, even among people of more than average intelligence, that milk as found in the market is very largely a mixture of chalk and w^ater. Upon what this absurd tradition is based, it is difficult to surmise, since even though a person were led to practise such a miserable fraud, he would discover that chalk and water will need constant stirring; to maintain even the outward semblance of milk, and that a few minutes' standing is sufficient for complete separation into a deposit of chalk and a fairly clear supernatant liquid. A less common, but equally absurd, notion that calves' brains are a common adulterant of milk, arose about half a century ago from the report of a microscopi- cal examination of a milk sediment in which certain particles were detected which bore a resemblance to nerve tissue. Calves' brains do not lend themselves readily to the making of emulsions, the supply is limited, and they find a fairly good market in their true character. Cane sugar is said to have been found at rare intervals, and gelatin is used occasionally as a thickening for cream. Starch is believed by many to be a common adulterant, but it is used very rarely. In the course of many years' supervision of a large pubhc milk supply, during which several hundred thousand samples of milk were examined for adulterants of all sorts, but one instance of the use of starch fell under the author's notice. This was due to a shortage in the normal supply, which led a dealer to dispense a mixture of water and condensed milk, which latter component had been thickened with starch. CONDENSED MILK. Condensed milk is prepared by evaporating milk to about a third or a fourth of its volume in vacuum pans. It is sold in bulk for immedi- ate use, and in hermetically sealed tin cans for use as occasion demands. Most of that sold in tins is made from skimmed milk, and is, therefore, very deficient in fat ; and much of it contains a large proportion of cane sugar, which is added to increase its keeping qualities. Condensed milk is in many respects and under certain conditions a valuable food preparation, but its use in infant feeding when other milk is obtainable is not a wise one, since it is deficient in one of the most important ele- ments, and contains another which is not a normal constituent. KOUMISS AND KEFIR. These are fermented preparations containing lactic and carbonic acids and a small amount of alcohol. They are produced through the action 96 FOODS. of niicro-orgauisms v/hich induce fermentative changes and bring about a partial conversion of the proteids to albumoses and peptones. Both had their origin in Russia, where they have been in use for many years. Koumiss is made generally from the milk of mares, but may be made from that of cows with the assistance of added sugar. Kefir is made more commonly from the milk of cows. Both are efferv^escent liquids having somewhat the taste of butter-milk, and are valuable in the feeding of the sick and of those with impaired digestive function. The " kefir grains " are small, hard, granidar particles which contain the requisite organisms. They are added to the milk after being soaked until soft, and their action is completed in two or three days. CREAM. Cream, as already stated, may be defined as milk containing a large excess of fat, and correspondingly lacking in water. The degree of rich- ness is dependent upon the method employed in its separation from the original volume of milk. That obtained by the conmion method of skimming contains ordinarily about 16 to 24 per cent, of fat, while that se])aratcd by the centrifugal machine contains from 20 to upward of 50 per cent., according as the machine is regulated for '' light" or " heavy " cream. The latter is so thick as to give rise to a common notion that corn starch is used as an adulterant. This substance, however, is used rarely if ever in this way. (Jelatin is employed as an adulterant to a limited extent. A ])reparation largely advertised to the trade at one time as a "cream thickener" was analyzed by the author, and found to be a mixture of gelatin, borax, and boric acid. The common adul- terants of cream are ]ireservatives and coloring agents. The former are used mostly during the hot months ; the latter during the winter, when, on account of the difference in feed, the cream has not the char- acteristic yellow tint so highly prized. Milk as a Factor in the Spread of Disease. Milk may act as a carrier of disease or cause of functional disturb- ance through infectious or poisonous matters originally present, or re- ceived or evolved during handling and distribution. Thus, milk may be poisonous by reason of matters derived from the feed or of sul)- stances formed after it is drawn ; it may contain organisms of various kinds connected with bovine diseases ; it may become contaminated in various ways with matter containing the exciting cause of various human diseases. Poisonous Milk. — Certain plants eaten by cows may cause milk to become unfit for drinking because of toxic properties. Poison ivy i^Rhus foxioo(Jc)tdron), for exam]ile, causes in cows a condition known as "trembles," during the continuance of which their milk is .said to cause severe gastric symptoms with great weakness. The most ]iromi- nent symptoms are pain, nausea, vomiting, constipation, and subnormal temperature. If the milk be boiled, the poisonous properties are de- MILK AS A FACTOR IN THE SPREAD OF DISEASE. 97 stroyed. According to Dr. D. D. Grout/ whatever the nature of the poisou in milk from cows afflicted with " trembles," it attacks the cen- tral nervous system and produces characteristic trembling and profound loss of muscular power. He believes that a peculiar microzyme exists in the blood, and has pathogenic properties, which may be reproduced indefinitely through the milk and through butter and cheese made therefrom. The leaves of the common artichoke are said ^ to cause abdominal pain, vomiting, and diarrhoea. As stated on a preceding page, milk may undergo a peculiar form of decomposition residting in the production of an intensely violent poison, a benzene derivative, knoA^m as tyrotoxicon. Fortunately, this is an uncommon change, but it betrays to the consumer no sign of its occurrence at the time of drinking. The effects produced are various, and are well illustrated by the following cases : Case L— Reported by Dr. W. K. Newton and Mr. S. Wallace.^ On August 7, 1886, 24 guests of one hotel at Long Branch, and 19 of another hotel at the same place, were taken sick soon after supper with the same train of symptoms, which were nausea, vomiting, cramps, and collapse, dryness of the throat, and burning sensation in the oesoph- agus ; in many cases there was absence of diarrhoea, and in several there was active diarrhoea without vomiting. Many had violent vom- iting followed by collapse. As a rule, the nausea and vomiting were persistent and obstinate, and accompanied by a tendency to exhaustion and collapse. A week later, 30 guests of still another hotel were seized in precisely the same manner. The onset occurred in from one to four hours after eating, but in one instance the symptoms appeared almost immediately after drinking about a quart of milk. Investigation showed that the trouble was due wholly to milk, for only the milk-drinkers were seized, and those who had had no other food were the worst sufferers. The three hotels were served by one dealer, who made two deliveries daily. The milk of the second delivery was the cause of the mischief in each outbreak. It was drawn at noon, and, without being cooled at all, was carted eight miles in the heat of the day. The cows were healthy and Avell fed. In a portion of the milk that caused the third group of cases, the presence of tyrotoxicon was demonstrated. Case II. — This was a most extraordinary outbreak, limited to a family consisting of father, mother, son, and daughter, of whom all but the first mentioned died. The family physician called Professor V. C. Vaughan in consultation after the fourth member of the family was seized, and from his report of the case the following facts are taken. The first one seized was the father, a man of fifty years. When first seen, he was vomiting severely, his face was flushed, and his tem- perature was subnormal (96° F.). There was marked throbbing of the ^ Quoted in American Medicine, Aug. 31, 1901. 2 Milchzeitung, 1891, p. 40. 3 Medical News, September 25, 1886, p. 343. 98 FOODS. abdominal aorta, the tougue was heavily coated, and the breathing was very labored. The pupils were dilated, and much of" the body was cov- ered with a rash. The vomiting continued some hours, the vomirus being colored with bile. The bowels had not moved, but under the influence of a cathartic, a stool occurred on the following day. Retching and vomiting continued during that day and niglit, and there was persistent stupor. During the following three days, there Avas but little change. Then improvement began, but recovery required a month. The son, a strong youth of eighteen, was the next to be seized, four days after the beginning of his father's sickness. The symptoms were similar, but were more violent, and there was no rash. On the following evening, the mother, about forty-five, was seized in the same way, and on the succeeding evening, the daughter also. On the day following the last seizure, none of the cases showed improvement. The temperatures were subnormal, 94° and 95° F. All complained of a burning con- striction in the throat and difficult swallowing, and called frei|uently for ice. Two days later, the mother and son died ; the daughter grew worse, became unconscious, remained so three days, and then died. Post-mortem examination in the case of the daughter revealed no characteristic lesions to account for death. The outbreak was most carefully and thoroughly investigated from every standpoint, and the conclusion reached was that tyrotoxicon was the cause. The milk had been kept in a buttery which was in a most unsanitary condition. During three years, the family had suffered frequent attacks of like character, but they were much less severe. Fresh milk, placed in the butterv over night, and then examined for tyrotoxicon, gave unmis- takable chemical and physiological evidence of that j)(>ison. Fresh milk inoculated Avith dirt from the butterv floor also developed it, as did also other portions treated with Aomitus, stomach contents, and aqueous extract of the intestines, while a fifth s])ecimen untreated remained free from it. All the evidence in this case pointed to the more or less constant ])resence of ])oison in the milk, and the AA'ide variation in the time of seizure in the final outbreak indicates that all Avere not affected by the same day's supply. Milk from Diseased Cows. — The milk of cows suffering from the ])rominent cattle plagues is more or less altered in composition, and there appears to be evidence that it may be actually dangerous. In rinderpest, the proteids are much increased — in fact, more than doubled ; the mineral constituents are considerably increased, and the fat and sugar are diminished. \\\ foot , found by the same investigator in 3 of 8 samples from individual cows of first-class dairies ; by Leblauc,^ in the ducts of 10 of the 24 teats of 6 daily cows examined ; and Staph nlococcus aureus and a/bus, found by Leblanc, v. Hellcns, and others. B. coli communis is present almost invariably in milk from all sources, and Proteus vulgaris is found fre- quently. The presence of the various pyogenic bacteria in milk, whether due to their existence within the milk ducts or to even slight lesions on the hands of the milkers, is a matter of grave concern as a common cause of serious gastro-intestinal disorders, especially in children in their first years. It would appear from a report made by GafFky,- that the milk of cows suffering from specific enteritis may be a cause of sickness. Three persons connected with the Institute of Hygiene at Giessen were seized, after drinking milk from a cow suffering from such a disease, with nausea, vomiting, diarrhwa, and mental confusion. One recovered in a few days, the others in about four weeks. The milk was drunk in the raw state. Concerning the agency of milk of tuberculous cows in spreading tuberculosis, there is, as m the case of tuberculous meat, a wide diver- gence of opinion. There can be no doubt that the milk of such cows may convey the infection to other animals, but whether to man camiot be definitely stated, because of the impossibility of experimentation, and since, in any case of supposed transmission, very many other possible agencies must be eliminated. As stated on a preceding page, local in- fection by meat through wounds incurred at autopsies of tuberculous animals is not impossible, but cases of similar infection through milk are exceedingly rare. Salmon ^ cites but 3 cases in all ; one, from the application of cream to a leg supposedly poisoned by ivy ; a second, from milking with a wound in one finger ; and a third, from attempted removal of tattoo-marks by the introduction of milk through needle- punctures. There can be no doubt that the tubercle bacillus finds its Avay into milk, particularly if the udder is involved, but even when not. This was asserted so long ago as 1889 by Professor H. C. Erust,* who, after a very extended inquiry, proved that the milk of cows with tubercu- losis in any part of the body, and with no local lesion of the udder whatever, may contain the bacillus. This finding has been confirmed by a number of more recent observations. Especially noteworthy is the investigation pursued by Drs. RabinoAvitsch and Kempner,^who obtained positive results from inoculation experiments on guinea-pigs Avith the milk of 10 out of 15 cows that had reacted to tuberculin. Of these 1 Lyon medical, 1900, p. o61. * Deutsclie niedicinisehe AVochensclirift, 1892, No. 14. •' Bulletin 33, Bnrenu of Animal Tndnstrv, 1901. * American .Journal of Medical Sciences, November, 18S9. ■' Zeitschrift fiir Hygeine und Infectionski-anklieiten, XXXI., p. 137. I MILK AS A FACTOR IN THE SPREAD OF DISEASE. 101 10 animals, only 1 showed clinical evidence of involvement of the udder, and onlv 1 other sho\ved any sign of it on microscopical exami- nation. Others who have obtained positive results from animals with normal udders include Bollinger, Delepine, Bang, and Adami. Similarly, the miUi of a tuberculous mother may be infective, even though no mammary lesions exist. Such an instance is reported by Roger and Garnier : ^ The woman died, seventeen days after confine- ment, Avith pulmonary tuberculosis. On the fourth day after delivery, 2 guinea-pigs were inoculated with her milk, one of them with positive results. The child lost weight from birth, and died at six months with tubercular lesions of the mesenteric glands, liver, kidneys, and spleen. But the c[uestion of excretion of bacteria by active mammary glands with no apparent lesions has not been studied exhaustively. According to Basenau,^ only those bacteria which are capable of acting on the walls of the blood-vessels so as to cause hemorrhages are able to pass from the blood into the milk, and in those cases in v/hieh B. tuberculosis has been detected in the absence of evidence of mammary lesions, the chances are that more or less alteration of the vessel walls has occurred in consequence of disturbed nutrition. The experiments of Basch and Weleminsky^ lead one to the conclusion that Basenau's position is correct. They infected animals with different species of pathogenic organisms, and found that, even when the blood teemed with anthrax bacilli, the milk showed no evidence of their presence i^iiiless there were local conditions esj^ecially favorable, such as vascular lesions, which may be caused by the hemorrhage-producing bacteria. It has been demonstrated by Ostertag* that the milk of cows which show no e\\- dence of tuberculosis beyond reacting to tuberculin contains no bacilli, and that calves and pigs fed thereon for months do not become tuber- culous. It is asserted commonly that the use of milk from tuberculous cows is a positive danger to public health, and attention is directed to the persistently high rate of mortalitv' from tuberculosis in all its forms among very yoimg children, and to improvement in the death-rates from other causes. It is asserted that this condition can be explained in only one way; that is, that a very large proportion of market milk is derived from tuberculous cows, and thus bottle-fed children, if at all susceptible, become infected. As to the probable proportion of infected market milk, owing to the wide differences in results obtained by various investigators, no definite statement can be given. Babino^^'itsch, for example, found it to be 28 per cent.; Massone^ by inoculation experiments placed it at 9 ; Ott® at 11.6. Sladen'' found that more than half of the samples taken from ^ Comptes rendiisde la Societe cle Biolosjie, March 2, 1900. ' Archiv fur Hyoiene, XXIII. (1895), p. 44. ^ Ibidem, XXXV. (1899), p. 205. ^ Zeitschrift fiir Hygiene und Infectionskrankheiten, XXXVIII. (1901), p. 415. " Annali d'Igiene Sperilnentale, 1897, p. 939. ^ Zeitschrift fiir Milch- und Fleischhvgiene, 1898, Xo. 8. ' The Lancet, January 14, 1899. 102 FOODS. the supply of the colleges at Cambridge (England) conveyed tubercu- losis to guinea-pigs on inoculation; but Eastes^ found the bacillus in but 11 out of 186 samples of milk which he examined. Others have obtained results anywhere within the range of 6 to 50 per cent. Doubt- less the differences are due to variations in local conditions, to differ- ences in technic, and to accidents always attending hap-hazard securing of any article of food in open market. Taking the mean of the figures given, and accepting that as a fair approximation of the extent to which public supplies are infected, it must be agreed that, if infection through milk is possible, the amount of disease so caused is quite small in proportion to the number of the population who are exposed daily to the danger. There are but few re])orted cases in which the mfluence of other possible conditions can be excluded so thoroughly as to leave no reasonable doubt of the causal relation of milk. Single instances are necessarily of less value than groups of cases, and the latter are much less conunon than generally is supposed. From the number available the following are selected as illusti'ations : BrouardeP records the death of 7 children with no hereditary taint, inmates of a convent, from tuberculosis supposedly induced by the use of milk from a cow with tuberculosis of the udder. Another case re])orted by him, and quoted by Freudenreich,-^ is one in which 5 of ] 4 girls in a boarding-school became infected and died. The milk which they had used daily came from a tu])ercul(ms cow. Demme* repoited as the only instance in his experience in which all other causes could satisfactorily be excluded, a group of 4 infants of healthy parentage fed upon uncooked milk of tuberculous cows. They all died of tuberculosis of the intestine, and the diagnosis was confirmed by autopsy. Later, he reported^ still another death from the same cause at four months. In this case also there was al)S()lutely no family history of tuberculosis. After the confirmation of the diagnosis by autopsy, the cow was slaughtered and found to be tuberculous. A more extensive outlii'eak among older children was reported l»y Ollivier to the Academy of Medicine, Paris, on February 24, 1S91. A Avoman of twenty-one years, of good family history, who had always enjoyed good health, died of tubercular meningitis shortly after taking u]) her residence in a boarding- house in Avhich within a short time previously 11 school girls had been seized with tuberculcisis. It Avas learned that the milk supply was derived from a single animal which was extensively tuberculous. Shortly afterward, still another girl died of phthisis pulmonalis in the same house. Some of these cases, if not all, may be accepted as very strong evi- dence that tuberculosis may be spread through the agency of milk. ^ British Medical Journal, November 11, 1S99. ^ Annales d'Hygit-ne piiblicjiie, XXIA'., ]>. (15. ^ Les Microtes et leur des Rolo dans la Laiterie, Paris, 1894, p. 4o. * .Jahresbericht iiber die Thiitigkeit des .lennei-'schen Kinderspitals in Bern, 1882, p. 48. * Ibidem, 1886, p. 20. MILK AS A FACTOR IN THE SPREAD OF DISEASE. 103 But if it is true that so large a proportion of the milk supply is from diseased cows, and that the disease is communicable m this way, it follows that with the vast majority' of di^inkers of raw milk the bacilli perish or are discharged without gaining entrance to the tissues. Granting that much of the public milk supply is derived from tuber- culous cows, and that it is consumed very largely in unsterilized condi- tion bv verv young children, one would naturally expect, if the bovine bacillus is markedly infective to man, to find a ver}' high death-rate from abdominal tuberculosis among the ^'ery young. It is asserted that this is the case, and elaborate arguments in favor of the statement that tuberculous milk is responsible for a great part of the constantly high infantile death-rate have been based on figm^es given by the late Sir E,. Thorne-Thorne, in his Harben lectures, in November, 1898, showing that, whereas in England and Wales the returns for 1891—1895, compared v,-ith those for 1851—1860, indicate a reduction in mortality from phthisis at all ages of 45.4 per cent., and from all forms of tuberculosis of 39.1 per cent., the decrease in tabes mesenterica was for all ages only 8.5, and for children imder five only 3 per cent. ; and that, moreover^ for children under one year there was not only no reduction, but an actual increase of 27.7 per cent. Such figures, emanating from so high an authority, would seem to admit of but one explanation, namely, that infected milk is a danger hardly to be overrated. But these figures are directly opposed to clinical experience elsewhere and, as will appear, are incorrect. Dr. D. Bovaird ^ points out that it is only in England that reports indicate any considerable number of cases of primary intes- tinal tuberculosis, and asserts that it is very rare in and about New York City, and that the evidence connecting tuberculosis in children with infected milk is veiy meagre. Koch has called attention to the great infrequency of primary tul^erculosis of the intestine among children in institutions in Berlin ; and Biedert," too, asserts that the amount of tubercular mfection through the alimentary canal is very small. Adami ^ is of the opinion that tuberculosis of young children, and especially peri- toneal and intestinal tuberculosis, is remarkably rare in the great cities of Xorth America ; but Jacobi,"' while admitting that primary tubercu- lar ulcerations of the intestine and primary tuberculosis of the mesen- teric glands are rare, holds that peritoneal tuberculosis is very common. Adami cites the mortalitj^ returns for jNIontreal for the year ended Jmie, 1900, showang that of 935 deaths from tuberculosis, but 4 were of children under fourteen, and 3 of these were from abdominal tuber- culosis in children under five years. Crookshank^ dissents from the opinion that abdominal tuberculosis of children is connected with in- fected milk, but believes that not sufficient consideration is given to the possibility of infection from human sources. The fallacy of Thorne-T home's figures has been pointed out by Carr, ^ Archives of Pfediatrics, Dec, 1901. ^ Berliner klinische "Wochenschrift, Xovember 25, 1901. ^ Philadelphia Medical Journal, February 22, 1902. * Xew York Medical Journal, January 25, 1902. ° The Lancet, ^S^ovember 2, 1901. 104 FOODS. Guthrie, Donkin, and others, an J all arguments based thereon must fall to the ground. In December, 1898, Carr' showed that the vast ma- jority of cases returned as tabes mesenterica were probably of maras- mus, due to gastro-intestinal catarrh. Guthrie" concluded from the results of 77 autopsies performed by him on tuberculous children that the disease begins far more commonly in the chest than in the abdomen, and that tal)es mesenterica as a cause of dciith in youno- children is practically unknoAvn or extremely rare. Donkin, who contends that the oriijinal significance of the term " tabes mesenterica " no lonwr holds, says : ^ " We all know that all kinds of intestinal and other dis- orders are constantly styled ' tabes mesenterica ' by those who fail to cure them." Notwithstanding the paucity of cases which offer strong evidence of a ciuisal relation between infected milk and the occurrence of tubercu- losis, and in spite of the now recognized differences between the bovine and human bacilli, the possibility of danger in individual cases cannot lightly be brushed aside. According to Theobald Smith, ^ it is quite possible that something interferes with the absoi^^tion of bovine bacilli, while allowing the human bacilli to jiass ; and while racial differences probably prevent the alisorption of bovine bacilli under ordinary circum- stances, and a few bacilli are harmless, there is danger if the digestive tract is flooded with bacilli from tuberculous udders. Ostertag' advo- cates the culling out of all cows showing clinical evidence of tubercu- losis (beyond reacting), and especially of all with lesions within the udder. Leblanc^ is of the opinion that the milk of tuberculous cows is dangerous, not on account of the bacilli, but on account of the toxins that it contains, for it has been proved to have toxic properties. Michellazzi has shown that such milk injected into tuberculous ani- mals causes a reaction, and that the milk of a tuberculous mother Avill in time jirove toxic to her child. Milk Contaminated from without with Organisms Related to Human Diseases. — ^lilk may become contaminated with infec- tive matter in various Avays. It may receive it from the hands, jier- son, and clothing of the milkers and others by whom it is handled, whether they are themselves sick or convalescent, or acting in the capacity of nurse or attendant for others ; it may acquire it from unclean vessels rinsed in polluted water, or from water with which it has fraudulently been mixed. Outbreaks traced to milk generally involve a considerable number of ])ersons, and a[)pear with some sud- denness. In fact, it is the simultaneous appearance of a large number of cases that draws attention to the water supply or milk as a common cause. Sporadic cases are rarely traceable to milk. On account of the danger of specific contamination of milk, no per- 1 The Lancet, 1898, II., p. 1662. '' Ibidem, 1899, I., p. 286. 3 British Medical Journal, October 14, 1899, p. 1040. * Medical News, February 22, 1902. ^ Loco citato. * Lyon medical, April 14," 1901, p. 561. 3IILK AS A FACTOR IN THE SPREAD OF DISEASE. 105 sou sick with or convalescent from infectious disease, and no person having to do with the care of the sick, or with the disposal of their excreta, or with the washing of their linen, should be allowed to handle milk intended for the use of others. Public authorities are rapidly becoming awakened to the importance of restrictive measures in this regard, and in many communities it has been made a criminal ofPence to fail to give notice of the presence of cases of infectious disease at the place of production of milk or among those engaged in its distribu- tion and sale. Diphtheria. — A large number of epidemics have been reported in which a positive connection with the milk supply appears to have been fairly well made out ; but so far as is known, there is no connection between any disease of the cow and that which we know as diphtheria, although a number of outbreaks of diphtheria have been reported as traced to garget. The specific organism of diphtheria may be intro- duced into milk from the discharges of persons employed in the hand- ling and distribution of milk before they have recovered thoroughly from the disease. Dr. J. W. H. Eyre ^ found the bacillus of diphtheria in samples of milk supplied to a large school where a number of cases of the disease had occurred. The organisms gave the usual character- istics, and no reason appears for doubting their identity. Schottelius ^ proved that the bacillus of diphtheria can grow very rapidly in raw milk, less so in sterilized milk at ordinary temperature, but very much better at 37° C. Inasmuch as the organism may per- sist for long periods after the disease apparently has disappeared, and may be present in the throats of persons in apparent health, it need not excite wonder when it is reported present in milk. Cholera. — Undisputed evidence of the connection between milk and Asiatic cholera is not very common. There is some disagreement as to the viability of the cholera organisms in milk; thus, Hesse ^ found that fresh, raw milk exerts a destructive influence on them ; that, in fact, they begin to die as soon as they are mixed with it. He found that they die at ordinary room temperature within 12 hours, and at incu- bator temperature in from 6 to 8 hours. The age of cultures, the nature of the culture media, and the addition of the latter to the milk with the bacteria, appear not to affect the result. Sterilized milk was found to be a better culture mediimi. Basenau "^ disagrees with Hesse. He found that uncooked milk does not kill the organisms in 10 hours, that they are active after 38 hours, and that up to the point of coagu- lation of the milk they increase considerably in number. He found that in polluted milk they remain active at least 32 hours at different temperatures (room temperature, 24° and 37° C), and that they remain active even after the milk has coagulated. Weigmann and Zirn ^ found that the length of time cholera bacteria ^ Britisli Medical Journal, September 2, 1889. 2 Centralblatt fiir Bakteriologie, Abth. I., XX., No. 25. ^ Zeitschrift fiir Hygiene and Infectionskrankheiten, XVII., p. 238. * Archiv fiir Hygiene, XXIII., p. 170. ^ Centralblatt fiir Bakteriologie, etc., 1894, No. 8. 106 FOODS. remain active depends upon the ratio they bear to the number of other organisms present, and that in order to survive for many hours they would have to be added to milk in exceedinoflv larg-e numbers. The evidence that cholera can be disseminated through the agency of milk is exceedingly lunited, and about the only ciise free from doubt is that recorded by Simpson/ who relates that 9 cases of cholera occurred suddenly on a ship ui the harbor of Calcutta, 10 men of whose crew had obtained milk from a native. One drank but little and escaped, 4 died of imdoubted cholera, and o were ver}' sick with diar- rh(ea. Eight others who used condensed milk only, and those who used no milk whatever, were unaffected. It Avas learned that the vendor had diluted the milk about one-fourth with water from a tank to which dejections from cholera patients had gained access and iu which the clothes of the jxitients were washed. Scarlet Fever. — In December, 1885, occurred what has become com- monly known as the Hendou outbreak of scarlet fever, due to a disease of cows, and since that time a number of other epidemics have been traced apparently to a common milk supply. In the Hendon case, a number of cows were or had been sick with an infectious eruption of the udders, and there can be no doubt that the disease under considera- tion was spread through the agency of milk coming from this dairy ; but other cows liaving the same disease caused no trouble, and the pos- sibility of contamination from human sources could not be excluded absolutely. A number of other outbreaks of the disease have some- Avhat doubtfully been ascribed to similar teat eruptions, but in no case is the evidence conclusive. On the other hand, there is undoubted evidence that the disease has manv times been spread by milk from farms where cliildren and others were sick with it. In tracing epidemics of this and other diseases to a common cause, there is always danger of lending too nnich importance to coincidence, and of coming thereby to unwari-anted conclusions. As an illustration, the following case may be citevn to be a careful, cleanly, honest dealer, and was, therefore, the very sort of man to attract the particular class whose homes were invaded. The children affected belonged to closely affiliated groups of playmates. Further investigation revealed the fact that the first case was of a lad whose family was not a customer of the suspected dealer, and that, immediately before taking to his bed, he had been playing with a number of those who were among the next to be seized. These in their turn had been associated with others, and so the infection had spread. Thus, what might have served as a most useful example of a milk-borne epidemic of scarlet fever fell to the ground, and the unfor- tunate dealer was absolved from responsibility. Typhoid Fever. — There can be no doubt that, in the spread of typhoid fever, milk plays a part only second in importance to that of drinking- water. A very great number of epidemics have been traced beyond a possibility of dispute to milk coming from farms where cases of the disease have occurred. The contamination is brought about by the hands of the milkers or other handlers, who, in addition, assist in nursing, or by the addition of infected water, or through washing pails, cans, and other vessels in such water. From time to time, tabulated analyses of outbreaks supposed to be due to contaminated milk have been published, but a very large pro- portion of the cases included are based on very insufficient evidence, sometimes exceedingly slight, such as that a cow had drunk from water into which drainage from the barnyard had had access. But within recent years, a niunber of extensive epidemics in this country and else- where have been traced with as much definiteness to the milk supply, as have others to the water supply, and with the same and only defect that the bacteriological proof has been lacking. As is the case when outbreaks occur from polluted water, when attention is drawn to the possible cause, the bacteriological evidence is no longer obtainable, the conditions having changed during the period of incubation. The State Board of Health of Massachusetts has traced a number of extensive epidemics to the use of polluted milk, but in no instance has the organism been found in the milk. In the city of Boston also, where the local authorities keep a constant eye on the reports of typhoid fever cases with particular reference to the possibility of dissemination through milk, a number of small outbreaks have been traced definitely to milk supplies derived from small farms where persons sick with the disease were nursed by those who had milked the cows and handled the milk, and in these instances also the bacteriological evidence is lacking. That the organism can retain its vitality in milk, and even in sour milk, has definitely been settled. Hein found the organism in sour milk at 13°-18° C. after thirty-five days, but not after forty-eight. Hesse has found it in sterilized milk after four months. Drs. Fraenkel 108 FOODS. and Kister/ having; reason to believe that the unusual amount of ty- phoid fever at Hamburg during the summer of 1897 was due in part to infected buttermilk, undertook the study of the question whether B. typhosus can exist in that fluid, concerning which point there had been more or less of conflicting testimony. Obtaining some samples, they first investigated the number and identity of the contained bacteria, and learned that, while the number varied widely, the species were always about the same. Finding no pathogenic organisms, they steril- ized specimens in test-tubes a half hour a day for three days, then planted the typhoid bacillus in them and kept them at diiferent tem- peratures ; on ice and at 22° and 37° C. Loops were taken from each from time to time and planted, and each yielded positive results. The specimen kept at room temperature was under observation nine days ; the others were not examined after the third. The specimens of fresh buttermilk containuig all its bacteria were planted and kept under the same conditions, and from them the same results were obtained. Yet there was this difference, that there was always a diminution in the number of the pathogenic organisms, and this was the more marked, and sometimes very rapid, with increasing temperatures. Cholera Infantum. — In every large community, it has become cus- tomaiy to expect as a normal condition a large death-rate among chil- dren with the advent of hot weather. This increased death-rate is limited very largely to the very early age periods and to children fed on cows' milk, and Avhile children of the poor are the ones most com- monly attacked, those of the well-to-do are by no means free. During the siege of Paris, the infant mortality Avas reduced to a half of its yearly average, although the general death-rate had doubled. This unusual condition was attributed, no doubt correctly, to the fact that mothers were obliged to nurse their infants when they could, on account of the great scarcity of cows' milk and other foods. The common milk bacteria are ordinarily harmless, but it appears that some species under certain conditions produce toxins in sufficient amounts to cause gastric and intestinal disturbances. According to Baginsky,- a large part of the annual amount of cholera infantum is due to these products (see under Garget, p. 99). Dr. E. AY. Hope* investigated over a thousand cases of autumnal diarrhoea, and found that, of 233 deaths of infants under three months, only 16 had not received other than their natural food. That is to say, the deaths among artificially fed children under three months of age were fifteen times as numerous as amonir those nursed. In no less than 22 per cent, of the whole number of fatal cases, other members of the household had suffered from diarrhoea. The most striking instance of the communicabilitv of the disturbance was that at an infants' home in which were 10 children under the age of five months, all in jierfect health. An infant of two months was admitted in July with vomiting ' Miinolienor niodiciiiisclic AVocliensrlirift, Febniarv 18, 1898. -' Berliner kliiiische AVoflioiiscluift, 1894, 2s os. 43 and 44. =• Public Health, July,, 1899. MILK AS A FACTOR IN THE SPREAD OF DISEASE. 109 and diarrhoea, and within a few days 6 of the other infants and the nurses were sick in the same way. The 4 other children were taken away at once. The admitted child and the 6 that became infected all died. The 4 that were taken away were saved. Bacteriological examination of milk has shown the presence of ex- tremely active organisms, including B. enteritidis sporogenes of Klein,^ which has been found by its discoverer in the ileum contents of chil- dren and adults with diarrhoeal conditions, but not in a condition of heahh. It has been found by Andrewes in the discharges of cases of sporadic diarrhoea of adults, and by Klein in three different outbreaks among the inmates of a smgle hospital. It is a common saprophyte found in sewage, in polluted rivers, and m manured garden soil, and is very commonly detected in milk, the use of which has not been fol- lowed by imtoward results. Under certain mlkno^vn conditions, it becomes highly pathogenic, and recent milk cultures are intensely virulent when inoculated subcutaneously in guinea-pigs. It is probable that to tliis organism was due an outbreak of milk- poisoning in Malta, described and investigated by J. Zammit.^ In one village, 5 families comprising 12 persons were seized with vomiting, charrhoea, and cramps, and 2 children succumbed. Post-mortem exam- ination revealed nothing except congestion of some of the viscera. Sub- sequently, in another village, 1 7 persons in 5 houses were attacked with severe gastro-enteritLs and collapse. The symptoms, which came on in all cases about three hours after drinking milk, included vomiting, diarrhoea, acute pain in the stomach and bowels, cramps in the extrem- ities, weak and irregular pulse, and cold and clammy skin. The per- sons concerned in both outbreaks obtained their milk from the same dealer, whose cans, which had a sour smell, yielded on bacteriological examination a bacillus having all the characteristics of the one men- tioned. Families which were supplied by the same dealer, but directly from the goats, showed no symptoms, and the goats themselves were free from disease. Andrewes"^ has described 3 much more extensive outbreaks, referred to above, due to the same organism, in one of ^vhich the offending food was found to be rice pudding made with millv. The first and second -outbreaks, in which no one article of food could be incriminated, in- volved respectively 59 and 146 patients ; the third involved 86. In all 3 outbreaks, the great majority of the attacks were mild, but in some of the more severe cases, the discharges contained mucus and blood. In all 3, the organism was found in the stools, and in the second, it was found in the milk given out on the previous day. In the third, it was impossible to obtain any of the milk, but the pudding made with it yielded the organisms, in spite of the heat to which the compound had been subjected during its preparation. It was found by direct experi- 1 Centi-alblatt fiir Bacteriologie, etc., XXII., Abth. I., Nos. 20 and 21 ; XXIH.. Abth. L, Xo. 13. ' BritisTi Medical Journal, May 12, 1900, p. 1151. ^ The Lancet, January 7, 1899' 110 FOODS. WATER FAT C.C.120. JK -iVi -3H ment that the interior of such a pudding did not attain a temperature above 98° C. during cooking, a temperature below that necessary for the destruction of the spores, which are among the most resistant known. Analysis of Milk. For ordinary purposes of determining the quality of milk, the pres- ence or absence of added water, and whether it has been robbed of its cream, a complete chemical analysis is by no means always necessary, since much may be learned from Fig. 2. simple inspection by means of the Fig. 1. n lactodensimeter and the lactoscope. The lactodensmieter (rig. 2), or lactometer, is merely a large hy- drometer with a stem graduated to show specific gravities ranging from 1.015 to 1.040. The lacto^scope, invented by Professor Feser, is an instrument desiirned to indicate the approximate fat content of milk. It consists of 1 glass cylinder, into the base of which a smaller cylinder of white glass, closed at the top and mounted on a metallic base, is fitted. The larger cylinder is grad- uated along the side ; the smaller one bears a number of black hori- zontal lines. The instrument is shown in Fig. 1. The jirinciple of the instrument is based ujion the fact that the opacity of milk is due mainly to the fat globules in suspension, and that, therefore, the richer a milk is in fat, the greater is its opacity, and the more it must be diluted to reduce the opacity to such an extent as to permit the passage of light. Feser's lactoscope. '81111611^" The method of use is as follows : Four cc. of the specimen are de- livered from a pipette into the cylinder through the opening in its uj)per end, and then water is added in small portions and thoroughly mixed by inversion of the instrument, the orifice being kept closed by the tip of the fi>refinger. As soon as the successive additions of water have reduced the opacity of the mixture to such an extent that the black lines on the white cylinder can be discerned so distinctly that they may be counted, the height of the liquid on the scale is noted and the per- A^^ALYSIS OF MILK. Ill ceutage of fat indicated is read. Four cc. of skimmed milk will re- quire so little water that, when the lines can be seen, the level of the mixture will be very low on the scale, while with rich milk it will be correspondingly high, and with cream the whole cylinder will be filled, and even then the lines cannot be made out. Control analyses show that the instriunent gives very fairly accurate results. Xeither of these instruments alone can be depended upon to indicate the true quality of milk, excepting in the case of samples which are either very good or very bad. The specific gravity alone is especially fallacious as a guide for the following reasons : The specific gravity of normal milk at 59° F. ranges between 1.029 and 1.034. The removal of cream causes it to rise ; the addition of water causes it to fall. A normal milk when robbed of its cream may show a specific gravity of 1.036, and then if a small amount of water is added, the gravity is brought down to 1.032 ; that is to say, within normal limits. Thus, a milk after being doubly treated so as to reduce its nutritive value, may show a normal specific gravity, and, on this test alone, be classed as pure. Nor is this the only objection to a system of inspec- tion of this most important food based upon the use of the lactometer, since milks exceptionally rich in fat have a specific gravity below the normal, and thus may be condemned as watered. The lactoscope alone is also not to be depended upon in all cases, since a milk which shows a normal content of fat may be one of con- siflerable richness in that constituent and extensively watered. Thus, a specimen containing originally 4.50 per cent, of fat may be watered very considerably, and yet show 3.75 per cent, by the lactoscope. By combining the use of both instruments, however, the fallacies of either are exposed. A normal specific gravity shown by the one and a normal fat content revealed by the other will indicate that, even if the milk has been tampered with, it yet possesses average richness. A normal specific gravity with a low percentage of fat will indicate skim- ming and watering ; low specific gravity with normal or low fat, water- ing ; and high specific gravity with low fat, skimming. Low specific gravity with very high fat will indicate unusual richness ; thus, cream has a very low specific gravity, due to its preponderance of fat. As a test of the accuracy of this process of examination, the author^ caused to be analyzed under his supervision 1.714 specimens which appeared by those tests to be of good quality, and of this number but 8 were found to have deviated materially from the statute requirement of 13 per cent, of total solids. Determination of Specific Gravity. — In taking the specific gravity by means of the lactodensimeter, the milk is mixed thoroughly, in order to insure homogeneity, by pouring from one vessel into another ; a cylinder of sufficient depth to allow the instrument to float freely is filled with the milk, and the instrument is carefully inserted, not dropped, down to the bottom, and then released. When it comes to rest, the reading of the stem at the level of the surface of the liquid is ' Thirty-fii-st Annual ILeport of the Inspector of Milk, Boston, 1889, p. 11. 112 FOODS. noted. It should be borne in mind that air bubbles are retained rather tenaciously by the milk, and tend to lower the density, and, therefore, in mixing the milk, too violent action must be avoided, and a short time should be allowed for the bubbles present to rise to the surface and escape. Inasmuch as the gravity varies with the temperature, and the instru- ment is graduated for 59° F., either the milk should be brought to that temperature, or a correction should be made accord- FiG. 3. ing to the deviation above or below that point. If the milk is colder, the reading will be too high, and, if warmer, too low. It ls more convenient to make a cor- rection for temperature than to heat or cool the speci- men to the normal point. The deduction of a half de- gree of gravity for each five degrees of temperature below 59°, or the addition of the same amount for each four degrees above 59°, will be found to be approximately accurate corrections. Determination of Fat. — For the accurate determina- tion of fat, several methods are in use, including the following : I. The Paper-coil Extraction Method. — This process requires strips of thick iilter-paper, free from substances soluble in ether and alcohol, about 6.25 by 62.5 cm., and a Soxhlet extraction apparatus. The most approved fotm of the latter consists of three separate pieces which fit together by ground-glass joints (see Fig. 3). The top and bottom pieces are, respectively, an upright Liebig condenser and a flask. The middle piece, which is the part in which the extraction process occurs, consists of a glass cylinder, closed at the bottom, from which a nar- rower cylinder with open end projects downward. The two cylinders are connected by a side tube which opens into the upper portion of each, and also by a siphon which opens from the side of the bottom of the large cylinder, extends upward, then turns upon itself, pierces the middle part of the wall of the lower cylinder, and terminates within and just below its lower end. When in use, the substance to be extracted is placed Avithin the upper cylinder, upon the bottom of which is placed a wad of absorbent cotton, which prevents the entrance of solid particles to the siphon tube, or it is con- fined within a cartridge of thick filter-paper which fits loosely within the cylinder. AVhen the cartridge is used, it is best to plug its open end Avith absorbent cotton, in order to pre- vent the escape of fine particles of the contaijied substance. The three separate parts arc joined together and then mounted on a water-bath. The ether or other extracting medium is contained in the flask, the exact weight of which has been determined. The heat of the Soxhlet extrac- tion apparatus. ANALYSIS OF MILK. 113 water-bath causes the ether to volatilize, and the vapor passes upward through the side tube into the extractor and thence to the condenser, where, coming in contact with the cold surface of the inner tube thereof, it condenses and falls upon the substance to be extracted. As the proc- ess continues, the condensed liquid accumulates and gradually rises until it reaches the bend of the siphon, which, when full, begins to act and discharges downward into the flask until the entire liquid is returned to its starting-point. During its accumulation, it acts ujDon the substance within the cylinder, and extracts more or less of the fat or other sub- stance, as the case may be, which is carried in solution mto the flask. The volatilization continues, and the process is repeated again and again as long as is necessary, and in this way the whole of the extracted matter is Anally within the flask, since, being itself non-volatile, it remains behind, while the liquid by which it is extracted is sent con- tinually on its errand. On the completion of the process, the ether is sent up again into the cylinder, and before it reaches the level of the siphon the flask is disjointed. The remaining ether is expelled cau- tiously, and the flask with its contents is placed in an air-bath, main- tained at 100° C, and dried until its weight is constant. The increase in the weight of the flask represents the amount of matter extracted. In determining the fat of milk by this process the method is as fol- lows : To one of the strips of filter-paper, made into a coil, a definite weight of milk, about 5 grams, is applied in either of two ways. A small beaker containing about the required amount is weighed and then the coil is thrust into it, kept there until nearly the whole has been absorbed, and then carefully withdrawn and placed dry edge downward upon a sheet of glass. The beaker is then weighed again, and the loss in weight, which represents the amount of milk absorbed, is noted ; or the beaker containing the milk and a small pipette is weighed, and then the necessary amount of milk is transferred to the coil from the pipette, after which operation the weight of the beaker, pipette, and the remain- ing milk is noted, and the difference set down as the weight of the milk absorbed. The coil is then dried in an air-bath at 100° C. for an hour or more, at the expiration of which time it is ready for insertion into the extractor. After it has been acted npon by the ether about a dozen times, the flask is detached and treated as above mentioned. After being allowed to cool, the weight is noted and the percentage of fat calculated arith- metically. Example. — The amount of milk absorbed by the coil was 4.950 grams. The increase in the weight of the flask was 0.173 gram. Then the amount of fat present in the sample is obtained by the equation, 4.95 : 0.173 : : 100 : x, wherein x = 3.49. 2. The Werner-Schmidt Method. — In this process, equal volumes of milk and hydrochloric acid, about 100 cc. of each, are mixed in a test- tube and boiled for about a minute and a half, or heated on a water- bath or steam-bath until the mixture is dark brown in color. It is then cooled, and the mixture shaken with 30 cc. of ether. When the 114 FOODS. two liquids have separated, the supernatant ether is withdrawn by means of a pipette or blown out with the assistance of a double tube such as is used in wash-bottles, the deliveiy tube extending into the ether layer very nearly as far as the Ime of demarcation between the ether and the acid mixture. The operation is repeated with several fresh smaller portions of ether, and the whole of the ether used is collected in a weighed flask. Then the ether is distilled oif, and the flask Avith its residuum of fat is heated to constant weight in an air- bath, cooled, and Meighed. The process may be shortened considerably by treating the milk in a graduated tube and, after thorough shaking with ether, removing an aliquot part of the latter by means of a pipette and evaporating to dryness. From the weight of this residue, the amount of fat in the whole volume of ether can readily be determined. Since the milk taken is measured, and not weighed, a correction must be made for gravity. Example. — Amount of milk used = 10 cc. Specific gravitv of specimen = 1.032. AVeight of milk used = 1.032 X 10 = 10.32 grams. Amount of fat found = 0.397 gram. Percentage of fat in the original milk = x in the c(piation, 10.32 : 0.397 : : 100 : .r ; x = 3.84. 3. The Babcock Centrifugal Method. — In this process, equal volumes of milk and sulphuiic acid are mixed in flasks of special design with narrow, graduated necks, and then whirled in a centrifugal machine for a definite length of time. On the completion of the process, the details of which are given below, the fat in a pure condition is within the graduated neck, and the percentage is read directly off. The kind of flask used is shown in Plate III. It has a capacity of about 40 cc. The graduated portion of the neck has a capacity of 2 cc. The details are as folhms : 17.6 cc. of the milk are measured l)y means of a pipette and introduced into the flask. Then an equal volume of sulphuric acid, specific gravity 1.800, is added, and the two liquids are mixed thoroughly by gentle rotar\' motion. Then the flask is placed in a centrifugal machine made especially for the j)urpose, and whirled for five minutes, at the exjiiration of which time hot Avater is added up to the beginning of the neck. The flask is whirled again for two minutes, and more hot water is added so as to bring the fat layer well up into the neck. After further whirling for one minute, the depth of the fat layer is determined by reference to the scale. This process gives sufliciently accurate results for all ])ractical ]iur- poses, and is in common use at experiment stations in this country. It is much used at creameries for determining the butter value of milk sent in from the surrounding country. The employment of sul])hurie acid having a higher specific gravity than that given, say 1.820, is objectionable in that it frequently hajipens that it is impossible to obtain a clear fat layer. The fat itself may be turned a very dark color, and the sugar of the milk may be attacked to such an extent that charred jjortions of it will separate and accumu- late within and beneath the column of fit, and so prevent a satisfiictory reading. If the acid used is weaker than 1.800, all the casein may not J PLATE III, ^9 |-7 4-3 - I I Babcock Flnsk, showing Fat in Neck. ANALYSIS OF MILK. 115 be held in solution, and portions of it may mingle with the fat and destroy the accuracy of the test. In Plate III. is shown the fat layer in the stem as it should be, free from alteration of color and from charred sugar and particles of casein. It will be observed that the line of demarcation between the water and the fat in the stem is very sharp. For cream, a flask with a much broader neck is employed. 4. The Babcock Asbestos Method. — In this method, the dried total solids obtained by the method described below (No. 2) are extracted in a Soxhlet extraction apparatus. Determination of Total Solids. — 1. Weigh into a flat-bottomed platinum chsh of about 2 inches (5 cm.) diameter, 5 grams of milk. Place on a water-bath for an hour and a half. Remove to a hot-air bath, maintained at 100° C, until its weight is constant. Cool in a desiccator and weigh. The difference between this weight and that of the dish alone represents the total solids of the amount of milk taken, and, multiplied by 20, expresses the percentage of total solids in the sample. If for any reason it is desired to use the total solids for ex- traction in the Soxhlet apparatus, the dish may be partly filled before weighing with fine, clean, dry sand, or with freshly ignited woolly asbestos. One objection to the use of the total solids in this way is that it is extremely cliificult to remove the whole amount from the dish, to the sides and bottom of Avhich a portion will adhere with great ten- acity, and can be removed only by burning. To obviate this diffi- culty. Dr. C. L. Spaulding has suggested lining the platinum dish with very thin tinfoil, which, after the weight of the total solids has been noted, is withdrawn with the sand or asbestos, and with it mserted into the extraction apparatus. Formerly, the residue of the milk dried in the dish alone without sand or asbestos was used for the determination of fat by the Wanklyn process, which consists in filling the dish with freshly distilled naphtha or with ether, and allowing it to act upon the residue and dissolve out the fat, several portions being used, after wliich the dish is dried again and weighed, and the loss in weight taken as the measure of the fat contained. Inasmuch as the solvent cannot penetrate the horny layer which forms on the bottom of the dish, not all the fat can thus be ex- tracted, and the figures obtained are ordinarily about 0.5 too low. 2. The Babcock Asbestos Method. — In this process, the milk is weighed into a cylinder of perforated metal or mto a filter-paper cart- ridge filled loosely with freshly ignited woolly asbestos, subjected to a temperature of 100° C. until the weight is constant, and then cooled and ^veighed. The gam in weight represents the total solids of the amount of milk taken. The cylinder may then be slipped into the extraction apparatus and used for the determination of fat. 3. Determination of Total Solids by Formula. — Knowing the correct specific gravity and the amount of fat, it is possible to determine fairly accurately the amount of total solids by the use of the formula of Hehner and Richmond. This formula is as follows: i^= 0.859 T 116 FOODS. — 0.2186 G, in which F represents fat, T the total solids, and G the figures of the specific gravity beyond the first decimal place. Example. — The specific gravity of a specimen of millv is found to be 1.030, and its fat content 3.95. Then applying the formula, we have 3.95 = 0.859 T— (0.2186 X 30), or 3.95 = 0.859 T — 6.558, or 0.859 T = 6.558 -r 3.95 = 10.508, and T=: 12.23. In other words, multiply 0.2186 by the figures expressing .specific gravity, add the percentage of fat to the product, and divide the result by 0.8o9. The formula may also be used to determine the percentage of fat, the specific gravity and total solids being known. Determination of Milk Sugar. — The amount of lactose may be determined either chemically or by means of the polariscope. 1, Method by Fehling Solution. — Reagents required: Solution A. Dissolve 34.639 grams of pure suljihate of copper in distilled Avater and dilute to a liter. Solution B. Dissolve 1 73 grams of potassium sodium tartrate (Rochelle saltj in distilled water, add 100 cc. of sodium hydrate solution of 1.393 specific gravity, and dilute the mixture with distilled water to a liter. In making a determination, 10 cc. of each solution are mixed in a boiling flask of about 300 cc. capacity. The amount of copper con- tained in 10 cc. of solution A requires for its reduction 0.050 gram of dextro.se, or 0.0667 gram of lactose. Peocess. — Into a porcelain evaporating dish of suitable size, dis- charge from a pipette 25 cc. of milk. Add three or four times as much water and heat to 40° C. Add acetic acid, a drop at a time, with con- .stant stirring, until the mixture separates into curds and a fairly clear whey. Transfer the whole to a graduated 500 cc. flask, and dilute with water to the 500 mark. Filter a portion through a dry filter, and use the filtrate for titration. Dilute tiie mixed reagents in the boiling flask with water and boil over a Bunsen flame. From a burette graduated in tenths, add the filtrate from the curds a little at a time, and continue the boiling after each addition. As the blue color begins to appear faint, the addition should be made cautiously, in order not to overstep the end reaction. As soon as the blue color is discharged completely, note the reading of the Ijurette. The calculation is e.xceedmgly simple. Since 0.0667 gram of lactose is required to reduce the copper in the reagent, it follows that that amount of the substance is contained in the number of cc. of the whey used, and the percentage is obtained by the application of the rule of three. Example. — The color is discharged by 24.3 cc. of the diluted whey. Then in the whole amount of milk taken the amount of sugar will be X in the ecjuatiou 24.3 : 0.0667.: : 500 : .r. x = 1.372. The amount of milk taken was 25 cc, hence in 100 cc. the amount would be 5.49, ANALYSIS OF MILK. 117 and this amount divided by the specific gravity gives the percentage by weight. Supposing the specific gravity to be 1.030, for example, the 100 cc. of milk weigh 103 grams, and the percentage of sugar will be X in the equation 103 : 100 : : 5.49 : x. x = 5.33. Inasmuch as the means of the first equation are constants, the reckoning resolves itself into dividing four times their product, 33.35, or 133.4 by the number of cc. used, and dividing this result by the specific gravity of the specimen. 2. Method of Polariscopy. — The determination of lactose and other sugars by means of the polariscope combines the advantages of accu- racy and of rapidity. The instruments in common use are of two kinds : those of which the normal sucrose weight, that is to say, the amount of sucrose which, dissolved in water and made up to 100 cc, will show 100 degrees on the scale when observed through a 200 mm. tube, is 26.048 grams, and those hi which it is 16.19 grams. Of the former, the Ventzke-Scheibler and the Schmidt and Haensch modifica- tion, and of the latter the Laurent instrument, may be regarded as types. The Schmidt and Haensch triple field, half-shadow instrument possesses the advantage of doing away with the matching of colors, and hence may be used by those who are color-bKnd, and even with those not so afflicted gives, on the whole, the most satisfactory results. Process. — Into a flask graduated on the neck at 102.6 cc. if the in- strument used is one of which the sucrose normal weight is 26.048 grams, weigh 65.95 grams of milk, or into one graduated at 101.6 cc, if it is one of the other class, weigh 40.99 grams, add 1 cc of solu- tion of mercuric nitrate of pharmacopoeial strength, shake well, and dilute with water up to the mark. Filter through a dry filter-paper, fill the 200 mm. observation tube, and note the reading of the scale when the field of observation is uniform. The reading divided by 2 equals the percentage by weight of lactose. The weights 65.95 and 40.99 represent twice the normal lactose weights of the respective types of instruments. The graduations 102.6 and 101.6 are adopted instead of 100 cc, since the dried precipitated curds from the respective amounts of milk of average specific gravity have a bulk equal to the excess over 100 cc. Determination of Ash. — The ash may be determined by igniting the residue obtained in the determination of total solids, provided no other substance has been introduced into the dish with the milk. The ignition should be conducted at a low red heat until the ash is perfectly white. Then the dish is cooled in a desiccator and again weighed. The difference between this final weight and the origmal weight of the empty dish represents the amount of mineral matter in the amount of milk taken. Or a larger amount of milk, say 20 grams, may be evaporated with a few cc of nitric acid and the residue ignited as above. Determination of Proteids. — Having determined the total solids, fat, sugar, and ash, the proteids may be reckoned by difference — that is, by subtracting the sum of the fat, sugar, and ash from the total solids, or they may be determined directly by the Kjeldahl process. 118 FOODS. Avbicli depends vipou the couversiou of the nitrogenous matter into ammonium sulphate, which then is decomposed by an excess of strong alkali, ammonia being set free. This is expelled by heat, condensed with the accompanying steam, and received in acid of known strength. The process is as follows : Into a Kjeldahl digestive Hask introduce a definite weight, say 5 grams of milk, about 0.7 gram of mercuric oxide, and 20 cc. of sulphuric acid of 1.840 specific gravity, free from nitrates and ammonium sulphate. Place the flask in an inclined posi- tion and heat below the boiling-point of the acid for from five to fifteen minutes, or until frothing ceases. Then raise the heat until the mixture comes to boiling, and continue the process until the liquid is clear and has a very pale straw color. This will require ordinarily less than an hour. Withdraw the lamp, and drop in, in small quantities at a time, permanganate of potassium, until, after shaking, the liquid acquires a permanent giccn or pur})lc color. This addition is not always or even usually necessary to secure complete oxidation, but since it is sometimes required, it Ls best to make it a part of the routine. Allow the contents to cool, and then transfer them with about 200 cc. of distilled water, plus sufficient for thorough rinsing, to a distilling flask of about 550 cc. capacity, fitted Avith a rubber stopper and a bulb tulie connected with a very long Liebig condenser, the delivery end of which is fitted with a glass tube bent at right angles, so that it may dip beneath the surface of the acid into which the distillate is to be received. Add a few pieces of pumice or granulated zinc, or about 0.5 gram of zinc dust, to jire- vent bumjiing, and 25 cc. of a 4 per cent, aqueous solution of sulphide of potassium, to prevent the formation of compounds of ammonium and mercury, which are not wholly decomposable by alkalies. Shake, and then add of a saturated solution of sodium hydrate, free from nitrates, sufficient to make the reaction strongly alkaline, pouring it down the side of the flask so as not to mix at once with the acid con- tents. Xext connect the flask with the condenser, mix the contents by gently rotating, and apply the flame. Distil, and receive the distillate in a vessel contaming 50 cc. of decinormal sulphuric acid. AVhen aliout 175 cc. have passed over, it may be assumed that all anunf)nia has been expelled, and then the distillate is titrated with decinormal alkali, using cochineal or methyl-orange as an indicator. From the difference in strength of the decinormal acid, the amount of ammonia is calculated, and from this the amount of nitrogen ; and this multiplied by 6.25 gives the total ]iroteids. Detection of Added Coloring Matters. — Annatto. — To al)out 100 cc. of milk in a cylinder about 1.5 inches in diameter, add a few cc. of sodium carbonate solution, to insure a strongly alkaline reaction during the examination, and then introduce a strip of heavy white filter-paper about 0.5 by oJ^ inches, and set the whole away in a dark place over night. If any annatto color is present, it will, through selective affinity, pass from the milk to the fibre of the paper, which thercliv acquires a salmon tint, the depth of which is digpendent naturally upon the amomit of the substance present. The strip is withdrawn from the PLATE IV. A. strip of Filter Paper Dyed by Immersion in Milk Colored with Annatto. B. Same after Treatment with Solution of Protochloride of T:n. PLATE V. Fig. 1. -/, VI- l- ,, ' . \, ■-o -""w Ly, V --^. -• .- vl. ^ •v' V ,, '' *' '^"^ 'l^ ^ V ^ .. V >• V -^ i^v^ Fig. Residues obtained in testing Milk for Caramel. Fig. 1, From uncolqred milk. Fig. 2. From milk containing caramel. ANALYSIS OF MILK. 119 milk, washed gently in running water, and laid upon a piece of paper of the same kind as itself. If so much as 1 part of the annatto sohi- tion in 100,000 is present, the strip will show a distinct salmon tint. On dipping the strip into stannous chloride sokition the color is changed to pink. Another method, by means of which all the color in the amount of milk operated upon may be concentrated in a form best adapted for preservation and for exhibits in court, is as follows : Coagulate from 100 to 150 cc. of the specimen by the application of heat and acetic acid, and separate the coagulum by straining through a piece of cheese- cloth. The coloring matter, being insoluljle in acid media, is precipi- tated with the curd, which, however, will show to the eye scarcely any indication of its presence. The curd is placed in a mortar and tritu- rated with 50—75 cc. of ether, which next is transferred to a stoppered separating funnel and shaken with 10 cc. of a 1 per cent, solution of caustic soda. ^^Tien the two liquids become separated, the latter, which now contains the annatto color, is drawn off into two porcelain or glass dishes about an inch in diameter, in each of which a disk of filter-paper is placed. They are then set aside in the dark and left over night. The disks are then removed and washed in fresh water. If annatto is present, they will have acquired a color varying in depth according to the amount of the dye in the sample. One disk is immersed in stannous chloride solution, the other in weak sodium car- bonate, and then dried and mounted on a white card. The colors yielded by a specimen of milk to which no unusual amount of the adulterant has been added are shown in Plate IV. Caramel. — Pour 125 to 250 cc. of the suspected sample into an equal volume of 95 per cent, alcohol, and filter. The filtrate, if not perfectlv clear, should be returned and passed through until it is quite free from turbidity. Any caramel present will be in solution in the alcoholic filtrate, and may modifH' considerably its color, which normally is yelloAvish or greenish according to season, the latter obtaining in sjjring and summer. To 100 cc. of the filtrate add 2 cc. of solution of basic acetate of lead, which \vill precipitate the caramel together with any remaining proteids, the precipitate showing a slight brownish color if caramel has been used in sufficient amount to bring about the improved appearance which is the object of its employment. Filter, wash with cEstilled water, and dry in an air-bath. According as the amount of caramel present is large or small, the horny residue on the filter-paper ^^-ill have a more or less deep chocolate tinge. The residue yielded by a pure milk will be either almost colorless, or vellow, or slightly iucliued to brownish, but not to chocolate color. The appear- ance of the two kinds of residue is shown hi Plate V. Caramel may also be shown if we proceed according to the second method described for the detection of annatto. The curd, after being- freed from the whey and triturated with ether, gives up to this solvent only fat and annatto. If caramel or anilins are present, the curd will appear brownish in the one case and more or less intenselv vellow in 120 FOODS. the other. If the curd is now shaken with hydrochloric acid, one of the following changes will be observed : If aniliu-orauge is present, the color becomes bright pink almost immediately ; with caramel it becomes gradually brownish blue ; if neither is present, the change is to blue. Anilin-orange. — See preceding paragraph. A more direct method is proposed by Lythgoe.' Place 15 cc. of milk in a porcelain dish and add about the same volume of hydrochloric acid (specific gravity 1.200). Agitate gently, to bring about thorough mixing and to break up the resulting curd into rather coarse lumps. If anilin-orange is ]>resent, the curd will be colored pink ; if none is present, it will be white or yellowish. Detection of Preservatives. — Borax and Boric Acid.— These sub- stances are detected easily either in the milk itself or in the ash after ignition of the residue. In the latter case, moisten the ash with a drop or two of strong sulphuric acid, and after a few minutes add 3 or 4 cc. of strong alcohol. Dip a strip of turmeric paper into the mixt- ure and allow it to dry without the aid of heat. In the presence of either of the substances sought for, the ])aper Avill have, when dry, the characteristic red color due to boric acid, instead of the yellow color which will be maintained in its absence. While the paper is drying, place the dish in a dark place and ignite the contained alcohol. If boric acid or its sodium compound is present, the flame will show at its outer edge a characteristic greenish coloration. This is shown most strongly directly after the alcohol is ignited. In the original milk, the test may be made in the following manner : Mix a few drops of the milk and an equal amount of fresh tincture of turmeric in a small porcelain dish and evaporate on a water-bath to dryness. ]\Ioisten the surface of the residue M'ith dilute hydrochloric acid, and dry again. If either of the substances is present, the residue will be light pink to dark red in color, and the addition of a drop of ammonia-water will change this to a green or greenish blue, according to the amount of the j)reservative present. Salicyclic Acid. — 1. Coagulate about To to 100 cc. of milk with mercuric nitrate solution or hydrochloric acid, and separate the whey by filtration. Shake the whey with ether, decant the ether into a watch-glass, and allow it to evaporate. To the residue on the watch- glass, ap})ly a drop of neutral ferric chloride. If salicylic acid is jires- ent, the characteristic purple coloration is ])rodu('ed. 2. jNIix the milk with phosphoric acid and strain through cloth. Place the liquid in a flask, connect with a condenser, and distil. Test the distillate with ferric chloride from time to time. Any salicylic acid present will go over with the steam, most of it toward the end of the operation. Formaldehyde. — ^lany processes for the detection of this substance in milk have been devised, some exceedingly simple and others (juite complicated. Those which give the best results and the greatest satis- faction are, on the mIioIc, those which are the simjilest in application ^ Report of Massachusetts State Board of Health for 1900, p. 647. i ANALYSIS OF MILK. 121 and require the least expenditure of time. The test should be applied within a few days after the addition of the preservative, since after a time it cannot be detected. 1. Method by Decolorized Fuchsixe. — Thi-ough a solution of fuchsine 1 : 500 pass a current of sulphurous acid gas, obtained by heating copper wire or foil with sidphuric acid, until the color is dis- charged. Preserve in a glass-stoppered bottle. To 10 cc. of milk, add 1 cc. of the reagent and let stand ten minutes. Add 2 cc. of strong hydrochloric acid and shake or stir briskly. The color which appears in the iirst instance is discharged completely by the acid if no formal- dehyde is present ; otherwise, a violet-blue tinge remains. If the amount present is large, the end color will be correspondingly intense. This method will detect the admixture of 1 part of formalin in 50,000 of milk. If the milk be distilled first, and the first part of the distillate treated with fuchsine solution, the test is delicate to the extent of re- vealing 1 part in 500,000. 2. Method by Phloroglucin. — Add to 10 cc. of milk in a test- tube 2 or 3 cc= of a 0.10 per cent, solution of phloroglucin and 5 to 10 drops of a 10 per cent, solution of sodium hydrate, and shake. In the presence of formaldehyde a gradual red coloration appears ; otherwise, no such change is observed. This test is said to reveal 1 part in 50,000, but such a claim appears, according to the experience of the author and others, not to be justified. 3. Method by Ferric Chloride. — Mix in a porcelain dish 10 cc. each of milk and hydrochloric acid (specific gravity 1.200) and 1 drop of ferric chloride solution. Heat and stir vigorously. If formaldehyde has been added, a violet color will appear before the boiling-point is reached, varying in intensity according to the amount present. This process is exceedingly delicate, and will detect 1 part in 500,000 in the fresh condition. 4. Method by Commerclil Sulphuric Acid. — This test is ex- ceedingly delicate and very easily applied. It cannot be performed with pure sulphuric acid, since the presence of a trace of iron is neces- sary. If one desu'es to use a pure acid rather than the ordinary com- mercial grade, the addition of a very small amount of ferric chloride will be sufficient. Take about 15 to 20 cc. of milk in a test-tube and pour about 5 cc. of the acid gently down the side so that it shall pass under, rather than mix with, the milk. Let stand a few minutes, and then note the color at the junction of the two liquids. If formalin is present, even in the slightest traces, a violet coloration appears at the liue of junction. In- asmuch as pure milk will show a somewhat purplish color when in contact with strong sulphuric acid, a color Avhich may readily be mis- taken at first for that due to formaldehyde, and since also the charring that occurs at the line of junction will often obscure the reaction, the process as originally recommended is somewhat faulty. The objections are removed, however, by diluting the strong acid with water so that its specific gravity is reduced from 1.840 to 1.700. The action of the 122 FOODS. stronger acid on pure milk is shown in Plate VI., Fig. 1, which shows the dark color due to charring and the purplish color, above spoken of, due to the same cause. In Plate VI., Fig. 2, is shown the appearance of the line of junction of ])ure milk and the diluted acid. It will be observed that the color produced is but a faint yellow. In Plate VI., Figs. 3 and 4, are shown the zones produced in milk containing formaldehyde in the pro- portions of 1 part to 25,000 and 1 to 50,000 by the use of the diluted acid. As may be inferred, the reaction is jn-oduced rather more slowly with the weaker acid. It is best to allow the contact to continue at least an hour before noting a negative result. 5. Luebert's^ Method by Potassium Sulphate. — Place 5 grams of coarsely powdered potassium sulphate in a 100 cc. flask and dis- tril)ute over it 5 cc. of milk by means of a pipette. Then pour care- fully down the side of the flask 10 cc. of sul])huric acid (specilic gravity 1.840), and allow the whole to stand quietly. If formaldehyde is present, a violet coloration of the potassium sidphate occurs within a few minutes, and gradually difl^uses through the entire liquid. If none is ]>rescnt, the mixture will at once assume a brown color, which rapidlv changes to black. This test is sensitive to 1 part in 250,000. Chromates. — Froidevaux ^ recommends dissolving the ash of about 10 cc. of milk in a few drops of water acidulated with nitric acid and, after neutralizing with magnesium carbonate, adding a few drops of test- solution of nitrate of silver, whereby a red precipitate, chromate of silver, is formed. As a control test, he recommends taking up another portion of ash with water acidulated with sulphuric acid, and adding little by little tincture of guaiacum. In the ]M*esence of chromates, an intense blue color is produced, which disappears very quickly. This process will detect 1 part in 50,000. Guerin ^ claims greater delicacy for the follo\nng method : To 5 or 10 cc. of milk add 2 drops of a 1 per cent, solution of sulphate of copper and 2 or 8 drojis of freshly prepared tincture of guaiacum. Pure milk gives a greenish color, while milk containing 1 part in 100,000 will give an intense blue, which reaches it>: maximum in a few minutes. Methods of Distinguishing between Raw and Cooked Milk. — To determine whether or not milk has been cooked, Carcano* recommends placing a few cc. of the specimen in a porcelain dish, adding a few dro])s of not too old oil of turpentine, warming gently, and then adding tincture of guaiacum. If the milk has not been boiled, a blue color apjiears ; otherwise, it does not. Du)iouy ^ gives the folloAving tests : 1. Guaiacol. Equal volumes of milk and a 1 per cent, solution of guaiacol in water are mLxed and then treated with hydrogen peroxide. ' .Tonrnal of the American Clu'iiiiial Society, September, 1901, p. 682. ^ .Jonnial de Phainiacie et de Cheinie, 1896, p. 155. •^ Chemiker Zeitung, 1S97, p. 174. * Gioniale di Faniiacia di Trieste, 1S96, p. 275. * Journal de Phannacio et de Cheiuie, 1897, p. 397. ! PLATE VI. Fig. 1. Fig. 2. Fig. S. Fig. 4. ~^ Fig. 1. Coloration Produced, by Concentrated Sulphuric Aeid, Sp. Gr. 1.840, in Contact with Pure Milk. Fig. 2. Coloration Produced by Sulphuric Aeid of Sp. Gr. 1.700 in Contact with Pure Milk. Fig. S. Coloration Produced by Sulphuric Aeid of Sp. Gr. 1.700 in Contact with Milk Containing 1 Part of Formaldehyde in 25,000. Fig. 4. Coloration Produced by Sulphuric Acid of Sp. Gr. 1.700 in Contact with Milk Containing 1 Part of Formaldehyde in 50,000. ANALYSIS OF MILK. 123 The immediate production of a yellow color indicates that the specimen has not been boiled. 2. Hydroquiuone. Three cc. of milk are mixed with 1 cc. of a fresh 10 per cent, aqueous solution of hydroquiuone and 15 drops of hydrogen peroxide. If the milk has not been boiled, a rose color im- mediately appears, and in a few minutes green crystals are deposited, 3. Pyrocatechin. Equal volumes of raw milk and an aqueous 10 per cent, solution of pyrocatechin are brought together and treated with hydrogen peroxide. With raw milk a yellowish-brown color is produced ; with boiled milk no color appears. 4. a-Naphthol. Raw milk gives with an aqueous solution of «-naphthol and hydrogen peroxide a violet-blue color. Boiled milk gives none. Storch's method^ is as follows : To 10 cc. of milk, add 1 drop of a 0.2 j)er cent, solution of hydrogen peroxide and 2 drops of a 2 per cent, solution of p-phenylenediamin and shake violently. If the milk has not been heated to 78° C (172.4° F.), an immediate blue color will appear; if it has been heated to 80° C. (178° F.), the blue color appears in about a half minute ; and if it has been heated higher than this, the blue will not appear at all. Sour milk should be neutralized with lime-water. Formaldehyde prevents the change to blue, but permits the occurrence of a faint red. The p-phenylenediamin solution keeps, in dark glass, about two months. Bernstein^ proposes the following: To 50 cc. of milk, add 4.5 cc. of normal acetic acid, shake gently until coagulation occurs, and filter. Heat the filtrate. If the milk has not been pastem-izecl, a hea\y pre- cipitate of lactalbumin will form. The higher the milk has been heated, up to 90° C. (194° F.), the smaller will be the precipitate; and if it has been heated beyond this, no precipitate at all will form. Detection of Gelatin in Cream. — For the detection of gelatin in cream, to wdiich it sometimes is added to give it body, Stokes ^ recom- mends the following procedure : Dissolve a quantity of mercurv in twice its weight of strong nitric acid (specific gravity 1.420) ; dilute with water to 25 times its bulk. To about 10 cc. of this solution add a like quantity of the cream and about 20 cc. of cold water. Shake the mixture vigorously, let stand for five minutes, then filter. If much gelatin be present, it will be impossible to get a clear filtrate. To the filtrate, or to a portion of it, add an equal bulk of a saturated aqueous solution of picric acid. If gelatin be present, a yellow pre- cipitate will immediately be produced . The whole operation is performed in the cold, and if the mercurv solution is ready, the test will not take more than ten minutes. Picric acid will show the presence of 1 part of gelatin in 10,000 parts of w^ater. ' Zeitschrift fiir Untersuchung der Nahrungs- und Genussmittel, 1901, p. 898. ^ Zeitschrift fiir Fleisch- und jSIilchhygiene, 1900, p. 80. ^ The Analyst, December, 1897. 124 FOODS. BUTTER. This valuable milk product is the result of violent agitation of cream until its fat coalesces into granular ]»rticles, which are then separated from the residual buttermilk, '' worked " to expel as much of the latter as possible, and, with or without the addition of salt and coloring mat- ter, formed into " prints " or " pats," or packed in bulk in boxes and firkins. Its natural color varies with the season, the so-called June butter, made when the cows from whose milk it is produced are feeding on grass, being bright yellow, while that made when they are stalled, and fed on hay and other winter feed, being almost white. The popu- lar demand being for a yellow article the year round, it is customary to secure this color out of season by the addition of annatto and other harmless vegetable coloring agents, the use of which has almost uni- versally the sanction of law. The flavor is influenced much by the character of the feed, by the care exercised in manufacture, by the amount of added salt, by age, and by the conditions of storage. Like milk, it absorbs odors very readily, both tht)se \\hich improve and those which impair its flavor. Taking advantage of this fact, it is the custom in the valley of the Var and in some other localities to place the freshly made product in prox- imity to jasmine, violets, tuberoses, and other flowering plants, in order that their fragrance may be absorbed. This practice is known as " enfleuraffc." The most delicatelv flavored butter under natural con- ditions is that to which no salt has been added, but it has the disad- vantage that within a short time it acquires a " cheesy " flavor, due to decomposition jn-ocesses. Owing to its lack of keeping qualities and to the very general preference for a more pronounced taste, the addi- tion of salt in varying amounts is the rule. Butter of good quality has but slight odor, but that which has undergone the conmion changes due to bacterial action has the characteristic odor and taste of rancidity. This is due to decomposition of the small amount of cnrd -which is en- tangled in the making, and which cannitt wholly be excluded. The fat itself, when se])arated from the curd by melting, keeps unchanged for long periods. In rancid butter, butyric and other acids are liberated, and others, as formic, are formed by absorption of oxygen. Under some unusual conditions not wholly understood, butter, without becom- ing rancid in the usual sense, undergoes a change to a perfectly white substance with a marked tallowy odor. Butter varies considerably in composition, but a fair average may be stated as follows : Fat ' 84.00 AVater 12.00 Curd 1.00 Salts 2.50 Lactose 0.50 It may be uinde to contain a much higher percentage of water, with correspon(Hngly less fat. \ BUTTER. 125 The fat is composed of gl vcerides of two groups of fatty acids, Avhich have been mentioned in the description of milk. Those of the in- soluble non-volatile acids, oleic, stearic, and palmitic, constitute about 92.25 per cent, of the whole; and those of the soluble volatile acids, butyric, caproic, caprylic, and capric, make up the remainder. It is to the second group that butter owes its distinctive flavor. The amount of water depends largely upon the thoroughness with which the buttermilk is worked out. In order that more water may be held, and thus a greater profit realized, some makers employ gelatin as an adulterant. One gram of this substance will take up about 10 grams of water, and, when mixed with butter in the right propor- tion, will hold water in the above ratio without affecting the consistence injuriously. Others employ glucose both for this purpose and as a preserv^ative. The salts include those natural to milk and those added for the pre- vention of rapid decomposition. The usual addition is common salt, but the use of boric acid and l^orax is extending gradually. Apart from the use of preservatives and of agents to assist in retain- ing water, butter is not much subject to adulteration, excepting in the sense that substittition of an article of less value when butter is called for is a form of adulteration. This substitute is known variously as artificial butter, butterine, oleomargarine, and margarine. Under the United States statutes, all butter or substitutes therefor made to resemble it, containing fats other than cream, shaU. be laiown as oleo- margarine. Following the origiual process, oleomargarine is made from fresh beef suet, which, after being cooled, washed, and ctit into very fine pieces by machinery, is subjected to a temperature of about 110° F. for several hours, in order to separate the fat from the tissue. It is then drawn off and kept for a time at 80° to 90° F., at which tem- perature the stearin solidifies, and then is separated by jjressure from the " oleo-oil." The latter is churned with milk or with milk and 2:en- nine butter, colored with annatto, and otherwise treated like butter. At the present time, oleomargarine is made not alone from beef suet, but to a much greater extent from " neutral lard," a j)roduct of leaf lard. Cotton-seed oil is used to some extent, but naturally it is not so well adapted to the purpose as the solid fats. Oleomargarine has been misrepresented to the public to a greater extent probably than any other article of food. From the time of its first appearance in the market as a competitor of butter, there has been a constant attempt to create and foster a prejudice against it as an unwholesome article made from unclean refuse of various kinds, a vehicle for disease germs, and a disseminator of tapeworms and other unwelcome parasites. It has been said to be made from soap grease, from the carcasses of animals dead of disease, from grease extracted from sewer sludge, and from a variety of other articles equally unadaptecl to its manufacture. The publication of a great mass of imtruth cannot fail to have at least a part of its desired effect, not solely on the minds 126 FOODS. of the ignorant, but even on tho.se of persons of more than average intelligence. So a prejudice was created against this valuable food product, but it is becoming gradually less pronounced. The truth concerning olecmiargarine is that it is made only from the cleanest materials in the cleanest possible manner ; that it is equally as Nvholesome as butter ; and that when sold for what it is and at its proper price it brings into the dietary of those who cannot afford the better grades of butter an important fat food much superior in flavor and keeping property to the cheaper grades of butter, which bring a higher price. Oleomargarine cannot be made from rancid fat, and in its manufacture great care must be exercised to exclude any material how- ever slightly tainted. Oleomargarine is not and cannot be made from fats having a marked or distinctive taste, and its flavor is derived wholly from the milk or genuine butter employed in its manufacture. It contains, as a rule, less water than does genuine butter, and consequently any difference in food value is in its favor. It undergoes decomposition much more slowly, and, indeed, may be kept many months Avithout becoming rancid. !Much has been said concerning its digestibility, and alarmists have gone so far as to claim that it is very indigestible, and likely to prove a pro- lific cause of dyspepsia, quite forgetting that the materials from which it is made have lield a place in tlie dietaries of all civilized peoples since long before butter was jH'omoted from its })osition as an ointment to that of an article of ftK)d. Many comparative studies have been made on this point, and the results in general have shown that there is little if any difference. H. I^iihrig ^ has proved by careful experiment that the two are to all intents and purposes exactly alike in point of digesti- l)ility. Oleomargarine has been the subject of a vast amount of restrictive legislation wherever it is made or sold. This has been passed in the interest of dairvmen and because of the ease with which it may be sold fraudnlently as butter at butter prices. To the practice of fraud in its retail sale, is due veiy largely the passage of proliibitive laws, many of which, however, have been declared unconstitutional. In Massachusetts, for example, it had at one time a very large sale, and in the city of Boston alone were nearly 200 licensed dealers. But the amount of fraudulent dealing was so great that the Legislature passed an act prohibiting its sale if it contained any ingredient causing it to look like butter ; in other words, no annatto or other substance which would cause it to be yellow could be used in its manufacture. Since its natural color is almost white, and since white liutter docs not apjieal to the eye, the result was practically the withdrawal of the article from ipen sale. In Germany, on account of fraudulent practices in the adulteration of butter with oleomargarine, the government passed, in 1897, a statute requiring the latter to contain 10 per cent, of oil of sesame, so that any subse<|ucnt admixture with butter may readily be detected by Bau- ' Zeitscluift fiir Untei'sucliung der Xahnmgs- und Cienussniittel, June, 1899, p. 4S4. BUTTER. 127 douin's reaction. This is a red coloration brought about when oil of sesame, fnrfurol, and hydrochloric acid are brought together ; and it is sufficiently delicate to show the adulteration of butter with 2.5 per cent, of oleomargarine containing the oil in the proportion stated. Experiment has shown that butter made from the milk of cows fed on sesame does not yield the reaction, but the fat of the milk of goats fed partly on sesame has been foimd to give it. The principal chemical difference between butter and oleomargarme lies in the relative amounts of glycerides of the soluble and insoluble fatty acids. Genuine butter-fat contains nearly 8 per cent, of butyrin, caproin, caprin, and caprylin, while the artificial product contains these glycerides only as they are introduced in the amount of milk or butter with which it is churned, for they are not present in suet, lard, and other animal fats. Of late years, high-grade butter has found another formidable com- petitor in what is known ^'ariously as renovated butter, process butter, and hash butter. The material from which this is made is gathered from dairies scattered over a wide expanse of country, and differs widely in color, texture, age, and flavor. It is melted, purified of its rancidity by washing, given the desired yellow color, and rechurned. Butter as a Carrier of Disease. — Since milk is known to be a car- rier of the germs of certain diseases under some conditions, the possi- bility that butter may act in the same way suggests itself, and the more strongly since, in ordinary creaming of milk, all but a very small pro- portion of the bacteria rise with the cream. Ordinary butter contains millions of bacteria to the gram, but whether the pathogenic forms can long survive has not been investigated very extensively, except in the case of the bacillus of tuberculosis. The bacteria of cholera and typhoid fever have been known to survive several days after being planted in butter, but beyond this we have little knowledge. Brusaferro, in 1891, produced tuberculosis in a rabbit through the injection of butter made from the milk of a cow with a tuberculous udder. Roth, in 1894, got similar results, and found, moreover, that 2 out of 20 market samples of butter used by him yielded positive re- sults. Schuchardt got negative results from 42 samples, while Ober- miiller found the bacillus in every sample of Berlin butter used in his first series of experiments. Dr. Lyclia Rabinowitsch ' examined 80 samples obtained partly in Berlin and partly in Philadelphia, and found genuine tubercle bacilli in not a single instance. She did, however, find a spurious organism, which produced in guinea-pigs changes which required very careful examination for the determination of its non- tuberculous character. It was present in 28.7 jDcr cent, of the samples. Petri- found it in 37.2 per cent., the genuine bacillus in 32.4 per cent., and neither the one nor the other in 30.4 per cent, Obermiiller,^ using salted butter in a second series, determined that the ^ Zeitschrift fiir Hygiene und Infectionski-anklieiten, XXYI., p. 90. ^ Arbeiten aus dem kaiserlicben Gesundheitsamte, 1898, p. 27. ^ Hygienische Eundschau, 1899, Xo. 2. 128 FOODS. injection of the butter-fat itself introduced a cause of irritation, and used, therefore, in his next set the watery fluid separated from the butter by heat and centrifngation. Four samples out of 10 from the same source as his tirst lot gave imdoubted evidence of the presence of genuine tubercle bacilli. Otto Korn ^ found them in 23.5 per cent, of samples purchased in Freiburg, and Dr. C. Coggi' in only 2 out of 100 samples purchased in Milan, though in a number of them the spurious organism was present. Morgeuroth^ has subjected oleomargarine to a similar investigation, since milk is used in its manufacture, and has re- ported positive results from 9 out of 20 samples. Annett* examined 28 sam])les of oleomargarine (15 from Berlin and 13 from Liverpool), and found virulent tubercle bacilli in only 1. We have as yet no evidence M'hatever that tuberculosis has ever been spread tlirough the agency of butter, but the subject deserves most thouo-htf ul consideration . Analysis of Butter. — Ordinarily, the examination of butter is lim- ited t<» the determination of whether or not it is mixed with or replaced by oleomargarine, but for the determination of its food value it is necessary to ascertain the proportions of fat and water. It is some- times of interest also to determine the amount of salt and the presence of othei- preservatives. Determination of Water. — Weigh a gram or two of the sample into a platinum dish, such as is used in the analysis of milk, and dry to con- stant M'eight on a water-bath. Determination of Fat. — Extract the residue from the preceding de- termiuation Avith ether or freshly distilled naphtha, being careful not to remove any of the particles of curd or salt. The process of extraction is very sim]ile, consisting in filling the dish about half full of the sol- vent and after a short time decanting it carefully into another vessel, and rejieating the operation until nothing is extracted. The solvent, or an aliquot part thereof, may be evaporated in a weighed beaker, or the dish may again be heated to a constant weight and the fat deter- mined by difference. The residue now represents the curd, lactose, and mineral matters. Determination of Salt, etc. — Ignite this residue at as low a tempera- ture as ])ossible, and thus burn off the casein and lactose. Their com- bined weight is ascertained by Mcighing the dish anew. What now remains in the dish is mineral matter, comprising the salts natural to milk and those added. Common salt may be determined by treating the final residue with water acidulated with nitric acid, and titrating in the usual way with standard solution of silver nitrate, using potassium chromate as an indicator. Another method of determining salt is as follows : Shake a known weight, 5-10 grams, of the sample with hot water in a stoppered sep- ' Archiv fiir Hvgiene, XXXVT., p. 57. 2 Giornale dolla R. Socicta italiana d'igiene, July, 1899, p. 289. •■' Hygionische Rundschau,. 1S99, No. 21. ♦ The Lancet, .January 20, 1900. BUTTER. 129 arating funnel until it is melted completely, let stand until the fat gathers on the surface of the water, and then draw off the latter through the stopcock. Repeat the operation with successive portions of about 20—25 cc. of hot water until a few drops of the washings, tested with silver nitrate, fail to show a cloudiness, due to silver chloride. Allow the combined washings to cool, and then, in an aliquot portion, deter- mine the chlorine by standard silver nitrate solution in the usual way. Determination of the Nature of the Fat. — To determine whether or not a specimen is or contains oleomargarine, an examination of the nature of the fat is necessary. As has been pointed out, genuine butter contains a considerable amount of volatile fatty acids, while the artifi- cial product contains very little ; but, on the other hand, the genuine article is correspondingly poorer in the insoluble non-volatile fatty acids. It is upon these differences in the two kmds of fat that the determination of the question of genuineness depends. The usual examination is limited to the determination of the volatile fatty acids in a given weight of the melted fat freed from water, curd, and salt. The fat is saponified, the resulting soap is dissolved in water and then decomposed by means of sulphuric acid, and the volatile fatty acids, thus freed from combination, are then distilled over, and their amount estimated by means of deciuormal sodium hydrate. Five grams of genuine butter-fat will yield an amount which will require at least 24 cc. of the alkali for complete neutralization, while an equal weight of oleomargarine yields so small an amount that, as a rule, less than 1 cc. is required. Mixtures give results between these limits, and from them one can estimate approximately the jsroportion of butter present. Process. — Heat a small piece of the sample on a water-bath in a suitable beaker until it is melted completely, and the contained water, salt, and curd have collected at the bottom. Decant a sufficient amount of the supernatant fat into a dry filter and allow it to pass into a shal- low beaker. When about 10 grams have been collected, place the beaker in a basin containing water and ice, and allow the fat to become hard. Place a small filter paper on one of the pans of the balance and counterbalance it exactly with weights on the other. Then weigh out as rapidly as possible 2.5 grams of the fat, transferring it to the paper by means of a spatula. Place the paper and fat in a 300 cc. Erlen- meyer flask, add 10 cc. of a 20 per cent, solution of caustic potash in 70 per cent, alcohol, and then place the flask on a water-bath. In a short time, especially with gentle rotation of the flask, the fat becomes completely saponified. Continue the heat until the alcohol is expelled, and remove the last traces of the vapor by blowing into the flask with a bellows or by swinging it in the air. Add 50 cc. of hot water, and when the soap is brought completely into solution, add 25 cc. of 10 per cent, sulphuric acid. The latter breaks up the soap, setting free both the soluble and insoluble fatty acids, the latter in the form of curds. Connect the flask with a Liebig condenser, after introducing several pieces of pumice stone to prevent bumping, and then, with the flask supported on a square of wire netting over a Bunsen lamp, distil 9 130 FOODS. slowly until 50 cc. have been collected. Titrate the distillate with deciuormal sodium hydrate, usiug pheuolphthalein as au indicator. AVith the amouut of fat taken, at least 12 cc. of the alkali will be required for neutralization, if the specimen is genuine butter. Many analysts prefer to employ 5 grams of fat and correspondingly larger volumes of water, and to distill 110 cc, whereof 100 is titrated. Some prefer also to cany on the process of saponification in a round- bottomed flask under pressure. Some measure the fluid fat directly into a weighed flask from a pipette, and ascertain the amouut taken by re- weighing after the fat has cooled and solidified. The saponifying agent is applied in diiferent forms, and many other variations in detail are recommended, but the end result is practically the same. The process described has been found in the experience of the author to be most satisfactory. The LeiFmann-Beam process has much to recommend it, particularly in the saving of time. The saponifying agent is prepared by mixing 20 cc. of 50 per cent, caustic soda solution and 180 cc. of pure concen- trated glycerin. To 5 grams of fat in an Erlenmeycr flask add 20 cc. of this solution, and then heat over a Bunsen flame until saponification is complete. This requires but a few nimutes ; the completion of the process is shown by the clear condition of the mixture. The soap is diluted with 135 cc. of bctiled water, added at first in very small amounts to prevent foaming. Then 5 cc. of dilute sulphuric acid (200 cc. in 1000) are added, and the preparation is ready for immediate distilla- tion. Distil 110 cc, mix thoroughly, and pass through a dry filter, titrate 100 cc, and to the result add -^^^ for the remaining 10 cc. If one Avishes to determine the amouut of insoluble fatty acids, it may be done in the following manner, but it should be said that the process requires much more time, and tliat the results are not always satisfac- tory, since the upj>er limit in the case of butter is so near the lower limit in that of olemargarine that samjiles yielding results close to the divichng line may need further analysis before an unqualified opinion of the nature of the specimen can be given. A mixture of genuine butter and oleomargarine may give results well within the normal limits of butter. Process. — Into a weighed beaker decant a few grams of the fat, and, when the latter has cooled, ascertain the amount taken by reweighing. Saponify as above described, evaporate the alcohol, dissolve the soap in water, and decom])ose it by the addition of an excess of acid. Heat until the prcci})itated insoluble acids are melted, then allow the whole to cool. When the fatty layer has assumed the character of a solid crust, break a small liole through it at a point on its circumference and another on the opposite side. Weigh a funnel and a dried filter of suitable size, ])lace the latter within the former, wet it thoroughly, and then filter the li(|uid from beneath the crust. Break uj) the crust, add boiling water, and transfer the whole to the Alter. AVash rei)eatedly with boiling water until the washings have no longer an acid reaction, then let drain until no more Water is discharged. The filter-paper BVTTER. 131 being wet, the melted fatty acids do not pass through with the wash- ings. Place the funnel and its contents in the beaker and dry in an air-bath at 100° C. to constant weight. The increase in the combined weights of the beaker, funnel, and paper represents the amount of in- soluble fatty acids in the amount of fat taken. Examination of Fat by Means of the Butyro-refractometer. — A sim- ple and quick method of ascertaining the nature of butter-fat with- out recourse to chemical analysis is that by means of the butyro-refract- ometer or other instrument designed for the purpose of measuring the refractive index. The instrument is shown in Fig. 4, with the prism Fig. 4. Zeiss butyro-refractometer. casing wide open. Its application recjuires so little time that, after a little practice, a person working alone can examine readily 15 or 20 samples in an hour. The method of use is as follows : The surface A and that to which it is opposed when the prism casing is closed should first be cleaned by means of a soft piece of linen moistened with alcohol or ether. Place the instrument so that the surface of the prism B is horizontal, then apply 2 or 3 drops of the clear fat, best fro-m a small filter paper held between the fingers. Close the prism casing and fasten it by means of the pin C The surfaces of the two prisms are now separated from each other only by the very thin layer of fat. With the instrument in its original position, the mirror D adjusted so as to illuminate the field clearly, and the upper part of the ocular so adjusted that the scale within is most clearly defined, read oif at what point of the scale the line between light and shade falls. Since the degree of 132 FOODS. refraction is influeucecl by the temperature, it is necessaiy to have some means of determining accurately the temperature of the specimen between the prisms. This is secured in the following manner : A cur- rent of Avarm water is conducted by means of a rubber tube connected with the inlet E into the prism casing, thence through the rubber tube F to the upper part, from which it escapes through the outlet G. The bulb of a thermometer projects into the curi-ent of water. The standard temperature for observations with this instrument is 25° C, and at this temperature natural butter, which has a refractive index of 1.459-1.462, will give a reading of from 49.5 to 54 on the scale, while oleomargarine, which has a refractive index of 1.465—1.470, Avill show 58.6 to 66.4, and mixtures of the one with the other will give from 54 upward, according to the percentage of oleomargine present. According to Wollny, to whom the invention of the instrument is largely due, any specimen which at a temperature of 25° C gives a higher rcadiup: than 54 will invariablv be found on chemical analvsis to be adulterated ; but he suggests that, in order to remove all chance of adulterated butter escaping detection, this limit be reduced to 52.5, iuid that all samples giving the latter reading be examined chemically. With temperatures other than 25° C, it is necessary to make cor- rections of 0.55 of a scale division for each degree C. The following table shows the maximum reading for pure butters at different tem- peratures : Temp. Sc. div. 1 Temp. Sc. div. Temp. Sc. div. Temp. Sc. div. 25° 52.6 31° 49.2 37° 45.9 43° 42.6 26 51.9 32 48.6 ! 38 45.3 44 42.0 27 51.4 1 33 48.1 ' 39 44.8 45 41.5 28 50.8 , 34 47.5 40 44.2 29 50.3 1 35 47.0 41 43.7 30 49.8 1 36 46.4 42 43.1 There are other processes for the investigation of the character of butter-fat, including the determination of the specific gravity, melting- point, iodine absorption number, and saponification equivalent ; but for all practical purposes the determination of the refractive index or of the volatile fatty acids is ordinarily sufficient, and the other determi- nations are merely corroborative. CHEESE. For thousands of years, cheese has been known as a veiy valuable food, and nuich attention has been ])aid to diiferent methods of manu- facture. At the present time, many varieties are made, their nature depending upon that of the raw material, the method of producing the curd, the ])roportions of the several constituents, and the method of ri])ening. Most varieties are made from cows' milk ; some are made fi'om that of ewes, and others Worn that of goats. CHEESE. 133 The milk is used either in its natural condition, or skimmed, or with the addition of cream. Generally, it is used in its natural condition. AVhatever the kind, the following is the general process of manufact- ure. The milk, with or without coloring matter as desired, is heated to 80° F. or above, and then curdled by means of rennet or by the acids formed by the ordinary milk bacteria. Usually, rennet is em- ployed ; sometimes, sour whey. The coagulation should be complete in from forty minutes to an hour. Too rapid coagulation causes the curd to be hard, tough, and unsuitable for the subsequent manipula- tion ; too slow action produces a soft curd difficult to work and not uniform in character. After the process of coagulation is complete, the curd is cut or broken into small pieces, and the whey is drawn off. Then the curd is gathered into a heap and covered, and allowed to stand for an hour or longer, during which time its increasing acidity assists in its hardening and promotes the separation of the remaining whey. When the curd has attained the proper consistence, it is placed in a cheese press and subjected to gradually increasing pressure, and after this process is completed it is removed to the curing place. For the proper ripening of cheese, it is essential that the curd be of the proper consistence throughout, and that only the favorable organisms be present, and these in not too great abundance. The curd produced by the action of sour whey is highly acid and inclined to be greasy. Owing to its high degree of acidity, it is not a favorable ground for the growth of many of the bacteria to which is due the production of the different kinds of flavor, and so the number of varieties possible of manufacture by sour whey is limited. Rennet, on the other hand, produces a curd which is elastic and not greasy or sticky, and which is a good culture medium for the bacteria whose assistance is needed. It acts best in milk which is slightly acid, for if the milk is neutral or only very slightly acid, the coagulation proceeds very slowly and the curd will not contract sufficiently to expel the whey ; if the milk is too acid, the process of coagulation is too rapid and the product too tough. A soft curd retains too much whey, and the fermentation of the milk sugar of the whey causes " huffing," or swelling, for the prevention of which, preservatives sometimes are employed. The bacteria concerned in the process of ripening exist in the original milk or in the air of the place of manufacture. Sometimes the varieties which produce cheese " faults " gain a foothold on the premises, and can be eradicated only by means of thorough cleaning and disinfection. The ripening process is carried on at about 70° F. It is essentially a process of decomposition, in which enzymes, bacteria^ and moulds are concerned ; and for the production of the same kind of cheese the same varieties of organisms must be present, and the particular variety producing a particular flavor must find the conditions such as are favorable to its predominance. It is not possible to start with milk that is entirely sterile, and then to inoculate with the par- ticular varieties wanted, since to sterilize milk completely requires the application of such a degree of heat as will produce changes in the 134 FOODS. casein, interfere ^vith the proper action of the rennet, injure the con- sistence of the curd, and destroy the enzymes. Ripening does not proceed satisfactorily when the curd has been produced through the action of acids. In ordinary' ripening, the casein is attacked by the organisms present, and ammonia, leucin, tyrosin, and several kinds of fatty acids are produced. The latter unite with the lime salts, which up to this }X)int have been in comljination with the casein. The acids formed include butyric and valerianic. From the lactose, we have, in addition, lactic acid. The process goes on at differ- ent rates with different kinds of cheese, and it may be short or long. In the production of certain forms of American and English cheeses, the individual specimens are sealed hermetically in tin boxes and kept at a favorable temperature for as long as four years, the boxes being turned each day. The ordinary grades of cheese, however, undergo comparatively short periods of ripening. Composition of Cheese. — The composition of cheese varies verj' much according to the nature of the raw material and the process of manufacture. The fat shows the greatest variation in amount, according as the cheese is made from whole milk, skimmed milk, or milk enriched with cre{\m. The most common American cheese is made from whole milk, as are also the leading varieties of English cheese, as Cheddar and Cheshire. The familiar Edam (Dutch) cheese is made from partially skimmed milk. English Stilton is a type of cheese made from milk enriched with cream. The cheese richest of all in fat is what we know as cream cheese, but, strictly speaking, this is not cheese at all, being simply fresh curd very rich in fat and not sub- jected to any process of ripening. The cheeses poorest in fat are those made from skimmed milk. They are tough, dry, and of but little flavor, and such as they have is inclined to be unpleasant. American cheeses of good quality may be said in general to contain about 36 parts of fat, 30 of proteids, 30 of water, and the remainder salts. The leading English cheeses, excepting Stilton, contain rather more water (about 35 per cent.), and correspondingly less fat. Swiss cheese has jiractically the stime composition, but contains rather more ])roteids and correspond- ingly less fat. Skimmed milk cheeses are particularly rich in proteids, containing often as high as 50 per cent. With the exception of those made from skimmed milk, it may be said in general terms that cheese is about one-third fat and one-third j)roteids. Of the many varieties of cheese put up in small bulk, mostly for use as a relish rather than as a substantial article of diet, the Ibllowing may be mentioned : Roquefort is made from the partly skimmed milk of ewes; it does not vary much in its ])erccntage of fat and proteids from American and English cheeses. Gorgonzola is very similar to Roquefort in composition and also in the method of manufacture. Both are npened with the assistance of moulds, which are mixed with the curd with the powdered bread crumbs on which they have been cul- tivated, and the cheeses are inoculated also after l)cing shape vast number of analyses of wheat show im]>ortant variations in the ])ercentage of its several constituents, for its quality is influenced considerably by climate, character of the .soil, and otlier conditions. According to H. W. Wiley ,^ a typical American wheat of the best quality should yield approximately the following results : Moisture 10.60 Proteids 12.25 Ether extract 1.75 Crude fiber 2.40 Starch, etc 71.25 Ash 1.75 100.00 These figures do not vary materially from the averages of a large number of analyses of samples of miscellaneous origin compiled by Konig, excepting in the proportions of moisture and starch, in which res]iects Wiley's tyjiical specimen shows superior value, being less rich in the one and richer in the other constituent. ^ American (jhemieal Journal, XV., j). 392. '■* Coniptes rendus, 1898, p. 12(i. 3 U. S. Department of Agriculture, Division of Chemistry, Bull. 13, Part 9, p. 1189. PLATE VII. O -•■'^, / .) O Wheat Starch. X 28S. Fig. 2. ^ O -P J / Rye Starch. X 285. WHEAT FLOUR. 139 Wheat Flour, In the manufacture of flour, the wheat kernels are subjected first to a process of thorough cleaning, and then are cracked, crushed, and ground until the required state of fineness is attained, the bran and other undesirable portions being removed by bolting. All flour is by no means the same in comj^osition and quality ; in fact, several grades of flour may be made from the same wheat by the employment of different processes of manufacture. Flours are graded according to color or appearance, those which make the whitest bread ranking high- est, although not equal in nutritive value to those classed as low grade. The flours of the several grades are known commercially as " patent," "family," ^^ bakers'," and other names which to the public have no special significance. Typical flours of the grades known as " high- grade patent " and " bakers' " should have, according to Wiley, approximately the following composition : Moisture. Proteids. Ether extract. Carbohydrates. Ash. Patent 12.75 10.50 1.00 75.25 0.50 Bakers' 11.75 12.30 1.30 74.05 0.60 The average composition of 210 samples of wheat flour of high and medium grades and of grades not indicated is, as given by Atwater and Bryant, as follows : Moisture 12.00 Proteids 11.40 Ether extract 1.00 Carbohydrates 75.10 Ash ■ 0.50 100.00 Thirteen samples of low grade averaged as follows : Moisture 12.00 Proteids 14.00 Ether extract 1.90 Carbohydrates 71.20 Ash 0.90 100.00 It will be noticed that the high grade flours are poorer in proteids and fat, and correspondingly richer in starch. Other grades of flour include those known as graham and entire wheat, Graham flour is understood generally to be a product containing all of the constituents of the original grain in their same proportions. AVhen it came first into use, such, indeed, it was ; but at the present time it is an unbolted] or partially bolted product of thoroughly cleaned and dusted wheat. Entire wheat flour is understood also to contain all of the original con- stituents of the grain, but is, in fact, made from wheat deprived of its outer coverings. It makes a somewhat dark-colored bread which is very palatable. Parenthetically, it may not be out of place to refer here to the absurd views maintained by a large part of the community as to 140 FOODS. the superiority, from a hygienic standpoint, of foods containing all of the constituents of the cereals from which they are prepared. It is difficult to understand how the nutritive value of any food can be increased by the retention of matters which are completely mdigestible and to a certain extent irritating to the digestive tract. It is argued that an all-wise Creator made wheat, for example, in the form m which we see it, and that it is not for us to attempt to improve it, as we think, by discarding the outer layers. But this sort of reasoning might be extended so as to favor the consimiption of the peel of oranges, the bones of fish, the feathers of birds, and other innutritions and undesir- able waste products. Preparations of Wheat Flour. — Bread. — First in importance of the preparations of wheat flour is bread. In the broad sense, bread includes all forms of baked flour, whether leavened or unleavened ; hi the common use of the term, it includes only those in whic-h leaven- ing agents are used, the other forms being designated as jiilot bread, crackers, biscuits, etc. The adaptability of wheat flour for bread-making is due to its gluten content. This substance, by reason of its tenacity, is capable of en- tangling the gas generated in the process, and by reason of its solidifi- cation by heat, furnishes a porous or spongy product easily penetrated and acted upon by the gastric juice. Not all cereals are capable of being made into bread, since, as will be seen, in most of them this very essential agent is lacking. For the ])rcparation of bread of good quality, the flour should con- tain not nuicli in excess of the average amount of moisture, and should be so cohesive that, after being compressed in the hand, it will keep its shape on being released. In the making of bread, the flour is mixed with >\'arm water or milk, salt, and yeast, kneaded into a stiff dough, and set aside in a warm place. The yeast attacks the sugar and splits it into alcohol and car- bonic acid gas ; the latter by its evolution and expansion causes the dough to become porous and to "rise." The fermentative process gives rise also to variable amounts of lactic and acetic acids. The raised douglv is then baked in suitable ])ans, and its porous character is increased by the further expansion of the gas by heat and is made ]iermanent by the solidification of the gluten by the same influence. If the fermentation is not allowed to proceed far enough, the resulting bread will be soggy or "heavy"; if too fiir, it will be sour. In place of yeast as a leavening agent, bicarbonate of sodium, com- monly known in the household as saleratus, and baking powders are employed very extensively. For the evolution of carbonic acid gas from sodium bicarbonate, the presence of an acid is necessary, and this is secured by the use of sour milk. First, the flour is mixed thor- oughly with the bicarbonate and tlien made into a dough with the milk. Bread made by this process is rarely of good quality, since it is difiicult to determine the proper amounts of the two agents for the best results, and any excess of the bicarbonate causes discoloration and disagreeable WE EAT FLOUR. 141 flavor, A better plan is to employ baking powder, which consists of sodium bicarbonate and an acid salt combined in such proportions that all of the available gas is set free from the alkaline salt and no unpleas- antly tastmg residue is left. The only advantage possessed by bakmg powders is the saving of time and labor ; the resulting bread is dis- tmctlv inferior to that made with yeast. The composition of the vari- ous classes of baking powders will be stated farther on. Another process of secui'ing leavening is that of spontaneous fermen- tation brought about by the enzymes present normally in flour. This process, Imown as " salt rising," is not in common use, requires much more manipulation than any other, possesses no advantages, and, there- fore, deserves no further mention. Freshly baked bread is much less digestible than that which has been kept a day or two. Its softness favors its clogging durmg mas- tication into a close mass which is attacked less easily by the gastric juice. In this country, however, it is the almost universal custom to eat bread, particularly in the form of breakfast rolls, not only in the fresh condition, but also hot from the oven. AVhen bread is kept for a day, it loses part of its moisture and acquires increased firmness and friability, which help maintain its porosity during mastication. Bread may acquire unwholesome properties on keeping, due to changes brought about in the presence of moisture by micro-organisms. Good bread is only slightly acid ; but if kept in a moist state, it is likely to become markedly so, and then may cause gastric derangement and diarrhoea in those not habituated to its use. Bread in this condition is undergoing fermentative changes that are hastened by the body tem- perature, with consequent evolution of gaseous products which cause flatulence and discomfort, and of irritating compounds which induce abdominal pain and diarrhoea. Bread made from old and partially spoiled flour is likely to have a distinctly sour taste and to be unwhole- some in the manner above described. ^Mouldy bread also is likely to be a cause of digestive derangement. Composition or Wheat Bread. — Since wheat flour itself is of vari- able composition, and since in the domestic manufacture of any article of food the processes employed are subject to slight or considerable variations, analyses of wheat bread necessarily must show great differ- ences in the proportions of the several constituents. Averages obtained from examination of samples of all sorts and of miscellaneous origin can hardly represent the composition of bread of good average or high quality. Wiley gives the following as the approximate composition of a " typical American high-grade yeast bread made with the best flour and in the most approved manner :" Moisture 35.00 Proteids 8.00 Ether extract 75 Ash 1.50 Fiber 30 Carbohydrates, other than fiber 54.45 100.00 142 FOODS. From analyses of bread made from three sorts of flour from the same lot of wheat, namely, "graham," "entire wheat," and "patent,'^ Professor H. Bnvder ^ shows that the nitrogen content is hip;hest in graham and lowest in patent flour ; but his digestion experiments prove that the latter has the highest proportion of digestible (available) protein. The lower digestibility of the protein of the others is due to the fact that both have a considerable proportion of that constituent in the coarser particles (bran), and that these resist the action of the digestive juices and escape digestion, so that the system derives less energy therefrom. Toast. — In the process of toasting, a large part of the moisture is dri\en otf, the surfaces are scorched, greater firmness is acquired, and the product is more easily digestible. Good toast cannot be made from perfectly fresh bread, on account of the moisture present, which causes sogginess ; it can be made only from bread at least a day old. The slices should not be thick, since then, while the surface is scorched, the interior acquires increased softness under the action of heat aud be- comes less digestible than the original bread. Rusks are much like toast. Instead of being subjected to the direct action of hot coals, the bread slices are baked for a time in a moderately hot oven. Pulled Bread is the crumb of freshly baked loaves pulled out in small masses and baked again like rusks. Crackers, or biscuits, are })re})arations made from unleavened dough and l)akc(l so dry as to be brittle. They keep well for a long time without losing their palatability. If not })ro})erly stored and cared for, they may, of course, become damp, musty, and mouldy. In composi- tion they varv" but little from the flour of which they are made ; they are drier, and what they lack in moisture they make uj) in fat, which, in the form of butter or lard, is added to ])revent them from becoming too hard and dry. Other preparations oi" wheat flour include cakes, which, on account of the contained butter, eggs, and sugar, are richer than bread ; pastiy, which, on account of its content of lard, is more difiicult of digestion ; and flour puddings, which, being very "close," require much time for digestion and often cause sensation of weight and op])ression. Macaroni, spaghetti, and vermicelli are preparations made w'xth hard wheat rich iu gluten. The flour is made into a stifi^ paste with hot water, and the comi^ound then is pressed through holes or moulds ui a metal plate and dried. They are exceedingly nutritious, but they are not as easy of digestion as other pre])arations of wheat, on account of their closeness. They were first made on a small scale in Sicily, but now are produced iu enormous amounts in Italy, France, Germany, and other countries. In their manufacture, American wheats are not held in higli esteem, containing not suflicient gluten and too much starch. The best wheat for the purpose comes from a particular district in • Bulletin No. 101, Office of Experiment ii>tations, U. 8. Departinent of Agricultuiv, 1901. WHEAT FLOUR. 143 Russia and from Algeria. Formerly, a grain from southern Italy was regarded as the most suitable. Adulteration of Flour. — Up to within comparatively recent years, flour has not been much subject to adulteration. Occasionally, certain mineral substances, as magnesium carbonate, gypsum, and ground chalk, have been reported in European samples ; but such have been employed as adulterants very rarely, if, indeed, at all in American flours. Alum has been added sometimes to flour of inferior quality to improve its color or to check beginning decomposition. Whether this addition is objectionable from a hygienic standpoint is a subject over which there is decided disagreement. It is believed by some that the amount of alum added is sufficient to exert an injurious effect on the digestive tract on account of its astringent action, and to bring about constipation and dyspepsia ; others believe that it can do no harm whatever, either to the consumer or to the nutritive value of the food ; and still others hold that, while it is not injurious to health, it lessens the nutritive value of the flour by forming insoluble aluminum phosphate, and thus depriving the system of the phosphates which otherwise would be absorbed. It is a fact that flour, treated with alum on account of begmniug deteriora- tion, has caused untoward effects, but it would be impossible to deter- mine how much influence should be ascribed to the alum and how much to the products formed by the fermentative processes in operation before the addition. The weight of evidence, however, is in favor of the view that alum is not incapable of producmg injury when taken into the system habitually in small amounts, and that it should be excluded from all articles of food intended for man. On accomit of the growing tendency to mix other mill products of inferior value with wheat flom-, such, for instance, as rye and corn flour, a law was passed by Congress in June, 1898, to meet the evil, and incidentally to make it a source of revenue. All adulterated flour is, by the act referred to, designated as " mixed flour," which term " shall be understood to mean the food product made from wheat mixed or blended in whole or in part with any other grain or other material, or the manufactured product of any other grain or other material than wheat." Under the provisions of the law, all persons engaged in the business of making mixed flour are required to pay a special annual tax, every package must be labelled plainly, the names of the ingred- ients being set forth, and upon every package of 1 9 6 pounds a tax of 4 cents shall be paid. Under the regulations of the Treasury, the term " mixed flour " is held not to include " the milling product from corn, rye, buckwheat, rice, or other cereals than wheat put upon the market as the flour or meal derived from such cereals, althougli the product may contain a percentage of wheat flour." The detection of other cereals and starches in wheat flour is accom- plished best by means of the microscope, since, as will appear, each has its characteristic appearance. According to Vogel, 70 per cent, alcohol containing 5 per cent, of hydrochloric acid remains colorless after being used to extract pure wheat or rye, turns pale yellow 144 FOODS. if barley or oats be present, and orange-yellow if mixed with pea flour. Rye. In external appearance, r>'e presents a close resemblance to Avlieat, but the kernels are darker in color and smaller in size. It is by no means so important as wheat as an article of food in this countiy, but in some parts of Europe it constitutes the mam food supply of the peiisantrv. Aecordino: to Wiley, a typical American rye should have approx- imately the following composition : Moisture 10.50 Proteids 12.25 Ether extract 1.50 Crude fiber 2.10 Starch, etc 71.75 Ash L90 100.00 American rye is smaller than that grown abroad, and contains less moisture. The proteids of rye are more like those of wheat than those of any other cereal, and in consequence rye stands next to wheat in adaptability for bread-makmg. The yield of gluten is inferior in amount to that obtainable from wheat. The starch of lye is much like that of wheat. The granules are rather more variable in size, the smallest of each kind being about equal, but the largest of rye somewhat surpassing those of wheat. There is but one point of difference in microscopic appearance which has any value in detecting the admixture of rye with 's^■heat, namely, that a certain fair proportion of the larger sized granules of rye jircsent irregular crosses or fractures. This is illustrated in Plato VII., Fig. 2. Bread made from rye flour is but little inferior in nutritive value to that from wheat, but it is less pleasing to the eye, being of a brownish tint, and it has a peculiar sour taste, not altogether agreeable on first acquaintance. Not uncommonly, its use by one not habituated to it causes a tendency to diarrhoea, which, however, is soon overcome. Barley. This important cereal is used mainly in the manufacture of beer, and but to a limited extent as a food. Deprived of its husk and rounded and jiolished by attrition, it is known as "pearl barley," and in this form is used more or less in the ]ireparation of barley-water, a drink for invalids. In its composition, barley is very similar to wheat and rye, but as its proteids yield no gluten, it cannot be made into bread. It is mixed sometimes Avith wheat flour for purposes of bread-making, but the product is less palatable and less digestible than ordinary bread. Wiley gives the following as the approximate composition of a typi- cal American unhulled barlev : PLATE Vlll. Fig. i: ;;-""V7' Jv^' -O; ■ C I ^ * #> Vr ^■Fc • c ^c/ '^;i^^ Barley Starch. X 28S. Fig. 2. ■^.-^ Oats Starch. X 28S. OATS. 145 Moisture 10.85 Proteids . 11-00 Ether extract 2.25 Crude fiber 3.85 Starch, etc 69.55 • Ash 2.50 100.00 The proteids inclucle, as in all cereals, a number of complex sub- stances, chief of which is hordein. The starch granules are like those of wheat, but are less variable in size. (See Plate VIII., Fig. 1.) In the manufactm-e of malt from barley for brewing, a peculiar nitroge- nous product, diastase, is formed, which has the property of convertmg starch to sugar. Oats. Oats are much used as human food in the form of oatmeal, which is the product of grinding the kiln-dried seeds deprived of the husk. The meal has a peculiar taste, which is both sweet and bitter. The composition of unhuUed American oats, as given by Wiley, is as follows : Moistm-e 10.00 Proteids 12.00 Ether extract 4.50 ' Crude fiber 12.00 Starch, etc 58.00 Ash 3.50 100.00 The mean composition of oatmeal, according to Blyth,^ is as follows : Moisture 12.92 Proteids 11.73 Fat 6.04 Sugar 2.22 Dextrin and gum 2.04 Starch 51.17 Fiber 10.83 Ash 3.05 100.00 The proteids of oats yield no gluten, and hence this article of diet cannot be made into bread, though with water it can be made into thin cakes, which are most palatable. Fat is present is greater abundance than in any other cereal. The starch granules are very small poly- hedra which show neither hilmn nor concentric rings. Thev tend to adhere together in masses of variable size, which are disintegrated easily bv trituration in a mortar. The single granules are sho"v\m in Plate t^IIL, Fig. 2. Oatmeal is a very nutritious article of diet, used largely as a break- fast food in the form of porridge. It has a somewhat laxative action, and, therefore, should not be eaten in irritable conditions of the bo^rel. It is also likely to disagree with some dyspeptics, because of its ten- dency to cause acidity and heartburn. ^ Foods: Their Composition and Analysis, London, 1896, p. 210. ]0 146 FOODS. Com. In the American usage of the word, corn includes the several varie- ties of Indian corn or maize. In England, the term is applied gener- ally to wheat, rye, oats, and barley, and more specifically to wheat ; in Scotland, it commonly means oats. In the United States, corn is in many ways the most important of the cereals, constituting in some parts of the countiy the chief bread food, and bemg the main sonrce of starch and glucose. The chief varieties are dent corn, showing a depression in the outer end of the kernel ; flint corn, having a hard smooth exterior ; sweet corn, rich in sugar and shrivelling Avhen ripe ; and pop-coni, a very flinty variety with small kernels, which contain a considerable amount of oil, which, in the process of roasting, explodes and causes the extru- sion of the starchy mterior in the form so universally familiar. The varietv in most common use, from whicli the several kinds of meal, hominy, and samp aro derived, is the flint corn. Hominy is the prod- uct obtained by grinding coarsely the kernels deprived of the hull by soaking. Samp is the whole, or practically the whole, of the kernel minus the germ and hull. Indian meal, or corn meal, is the product obtained by grinding the kernels between stones or by other processes of milling, and removing more or less of the bran by sifting or bolt- ing. According to the process employed, we have coarse and fine, and white and yellow meal. Prepared without removal of the germ, which is rich in oil, the product is ])rone to become rancid and mouldy on kee]iing. From a large number of analyses, Wiley deduces the following as the approximate composition of typical Indian corn : Moisture 10.75 Proteids 10.00 Ether extract , , , 4.25 Crude fiber , 1.75 Starch, et'; 71.75 Ash L50 100.00 The average of 19 analyses of samples of sweet corn by Clifford Richardson, quoted by AYiley, shows : Moistm-e 8.44 Proteids 11.48 Ether extract 8.57 Ci-ude fiber 2.82 Starch, etc 66.72 Ash 1.97 100.00 The composition of fine meal is given by Wiley as follows : Moisture 12.57 Proteids 7.1,3 Ether extract 1.33 Total carbohydrates 78.36 Ash 0.61 100.00 PLATE IX. Fig. 1. o o cP O *^ Q <^>^g G^ o "^ Co Corn Starch, x 285. ■V Sr^ Fig. 2. >^i%^ C Rice Slarch. ■; 285. RICE. 147 The lowered percentages of proteids and fats here sho^vn are due to the removal of the germ, rich in fat, and of the finer envelopes, rich in proteids. The proteids of corn, as determined by Chittenden and Osborne, are made up of several globulins, including myosine and vitelline, two classes of albumins, and two of zeins. The starch granules are poly- hedral with rounded angles, and have a punctiform, sometimes stellated, hilum. They are much larger than those of oats, which they resemble somewhat in form. They are shown in Plate IX., Fig, 1. On account of its deficiency in gluten, corn meal is not well adapted to the making of leavened bread, but it is used in many forms of sub- stitutes therefor. It is mixed with salt and water, sometimes with the addition of milk or eggs, and baked into not over-thick cakes, which, according to the method of preparation and baking, are known as johnny cake, corn dodger, corn pone, and corn bread. Sometimes, yeast and baking powder are employed. Corn meal is used extensively in the form of hasty pudding, or corn mush, and of Indian puddmg. In whatever form used, corn meal is a most nutritious and wholesome food. Rice. Rice is the principal food of a very large part, estimated at about a third, of the human race. Bemg, as will be seen, too poor in proteids, fat, and mineral matter to satisfy alone the needs of the body, the de- ficiencies are met by other vegetable products, as beans and peas, which are rich in these constituents. The form in which rice is seen in the household is the result of a polishing process which removes the reddish cuticle wliich the grain shows on removal of the husk. Wiley's figures, representing the com- position of typical polished rice, are as follows : Moisture 12.40 Proteids . . . . ; 7.50 Ether extract 0.40 Crude fiber 0.40 Starct, etc 78.80 Ash 0.50 100.00 Rice is the richest of the cereals in starch, and the poorest in all other respects. The proteids have not yet been studied systematically. Its starch is very easily digestible, and is very useful in all disordered con- ditions of the digestive tract when other solid foods cannot be borne. Under the microscope, the starch granules are seen to be much like those of corn, but are much smaller and have shai-per angles. They are separated less easily from one another, and are commonly in groups of variable size. They are shown in Plate IX., Fig. 2. Rice cannot he made into bread, but sometimes is mLxed with wheat flour, in order to give whiteness to the bread. It is used most com- monly in the freshly boiled condition or in the form of puddings. The 148 FOODS. most approved method of eookin_e: it is steaming;. This has the advan- tage of not taking away any of the already deficient proteids and saks, which to some extent are extracted in boihng, and also that it leaves the kernels distinct in themselves, and not aggregated in the form of a soggy mush, such as is produced often by improper boiling. Buckwheat. This valuable cereal is used very extensively in this country as a breakfast food in the form of pancakes eaten hot with syruj) or with butter and sugar. As it is devoid of gluten, it cannot be made into bread. The composition of typical American buckwheat is given as follows: Moisture 12.00 Proteids 10.75 Ether extract 2.00 Crude fiber 10.75 Starch, etc 62.75 Ash 1.75 100.00 The crude fiber is removed veiy largely in the milling, and is almost wholly absent from the white flour, a sample of which, analyzed by Wiley, had the following composition : Moisture 11.89 Proteids 8.75 Ether extract 1.58 Crude fiber 0.52 Starch, etc 75.41 Ash 1.85 100.00 Buckwheat is the most expensive of the cereals, and consequently is the most subject to adulteration with the cheaper members of the class. The admixture is detected readily by the microscope, since the starch granules have a \evy characteristic a])pearance, being small and angu- lar, and of nearly uniform size. Ordinarily they are seen in fairly large masses which are not disintegrated in the process of milling. The starch is shown in Plate X., Fig. 1. (6) LEGUMES. This group comprises peas, beans, and lentils. It is characterized by richness in proteids, which may be ]iresent in more than double the amount found in wheat. The chief proteid is legumin, which much resembles casein, and is known commonly as vegetable casein. Accord- ing to E. Fleurent,^ the proteids of this group consist of vegetable casein, composed of legumin and ghitenin, and vegetable fibrin, com- posed of albumin and gliadin. Thus : ' Comptes i-endus, 1898. PLATE X. Fig. 1. Buckwheat Starch. X 285. Fig. 2. ■ \ ^ *t..J' Pea Starch. X 285. PEAS. 149 ^T , , , • r les;umin 60.95 Vegetable casein I gj^^^^^^ • g^^^ T7 4. ui CI, • f albumin 0.64 Vegetable fibrin | gU^din 7.76 100.00 Their high conteDt of proteids makes them more satisfying than other vegetable foods, and enables them to act as a fair snbstitute for animal food. The millions of rice-eaters who, by reason of poverty or religious scruples, are denied the use of animal food, depend upon the legumes to supply the demands of the body for nitrogen. The East Indian, for instance, has no difficulty in satisfying his bodily needs with a hand- ful of beans added to his daily ration of rice. While legumes possess a very high nutritive value, they must be ranked as much more difficult of digestion than the cereals. They require prolonged boiling when cooked whole, but are prepared more quickly and digested more com- pletely when ground into meal and cooked with milk. Even under the most favorable conditions, a large part of the proteids is lost in the excreta. Rubner has shown that a fifth to a third is not digested and absorbed, whereas in the case of bread the proteid loss is less than a seventh. Some individuals are obliged to forego the use of peas and beans, on account of flatulence due to the formation of sulphuretted hydrogen from the sulphur in the legumin. This objection does not apply to len- tils, since they contain no sulphur. Peas. The average of 61 analyses of peas, compiled by Konig, is as follows : Moisture 14.99 Proteids 22.85 Fat 1.79 Crude fiber 5.43 Starch, etc 52.36 Ash . 2.58 100.00 When dried peas become old, no amount of boiling will make them soft, and they should then be soaked and crushed and cooked in some other way. The immature pea, so highly prized as a spring and summer vegetable, has a very different composition. Five analyses, compiled by Atwater and Bryant,^ yielded the following average results : Moisture 74.6 Proteids 7.0 Fat 0.5 Carbohydrates, including fiber 16.9 Ash 1.0 100.0 ^ Loco citato. 150 FOODS. The canned pea appears to contain considerably less nutriment. Of 88 samples reported by the same authorities, none contained less than 77.5 per cent, of water, and some contained as much as 92.7. Their average composition was as follows : Moisture 85.3 Proteids 3.6 Fat 0.2 Carbohydi-ates 9.8 Ash 1.1 100.0 The starch granules of peas are represented in Plate X., Fig. 2. Beans. There are many varieties of beans belonging to the two large groups, the broad beans and the kidney beans, but their composition is in gen- eral quite similar. Forty-one analyses of broad beans and 10 of kid- ney beans compiled by Konig give the following averages : Broad. Moisture v. 14.76 Proteids 24.27 Fat 1.61 Crude fiber 7.09 Starch, etc 49.01 Ash _3^ 100.00 Eleven analyses compiled from American sources by Atwatcr and Biyant yield averages not materially different. Five analyses of string beans in the fresh state and 29 of canned samples yield the following averages, showing, as in the case of peas, that the canned variety is less nutritious : Fresh. Canned. Moisture 89.2 93.7 Proteids 2.3 1.1 Fat 0.3 0.1 Total carbohydrates 7.4 3.8 Ash 0^8 _1.3 100.0 100.0 The Soja bean, which has been recommended highly in some quarters as a suitable food for diabetics, is remarkable for its high content of fat, and contains, in addition, so large an amount of starch as to make it quite tinsuited to the dietary of the diabetic. Konig has c(>n)])iled 21 analyses from all sources, and Jenkins and Wiiiton ^ have collected 10 more from American sources. The two groups give the following averages : ' Experiment Station Bulletin, No. 11, Washington, 1892. PLATE XI Fig. 1. Bean Starch. 28S. Arrowroot Starch. X 285. TAPIOCA. 151 xr;A«i™ Jenkins ^onig. andWinton. Moisture 9.51 10.80 Proteids 33.41 33.98 Fat 17.19 16.85 Crude fiber 4.71 4.79 Starch, etc . 29.99 28.89 Ash 5.19 4.69 100.00 100.00 Bean starch is shown in Plate XI., Fig. 1. Lentils. Lentils are the most nutritious of the legumes, but are not a popular food in this country, excepting among certain of the foreign-born population. Their use is, however, on the increase. The averages of 14 analyses compiled by K5nig are as follows : Moisture 12.34 Proteids 25.70 Fat 1.89 Crude fiber 3.57 Starch, etc 53.46 Ash 3.04 100.00 2. Farinaceous Preparations. Under this head are included sago, tapioca, and arrowroot. SAGO. Sago is derived from the pith of the stems of a number of species of palms. The pith is extracted and ground to a powder, which then is mixed with water and strained. The starch granules pass through with the water, and are deposited as a sediment, which constitutes the sago flour. From the flour, made into a paste, the various forms of granulated sago are prepared. Sago is an important starch preparation, and serves as a light and digestible food for invalids and dyspeptics, but its use is not restricted to these alone. It absorbs the liquid in which it is cooked, and becomes soft and transparent, but retains its original form. TAPIOCA. Tapioca is derived from a thick fleshy tuberous root called " mani- hot." The starch, which is extracted by a method similar to that em- ployed in the preparation of sago, is heated in a moist state on hot plates and stirred with iron rods, and thus forms irregular masses of transparent granules. In the process of heating, many of the starch granules become ruptured, and are then partially soluble in cold water. Tapioca, like sago, is useful for both sick and well. 152 FOODS. ARROWROOT. Arrowroot is a pure form of stareh from the tuberous root of the marauta. Its name is derived from the fact that the maranta root is believed to counteract the effects of arrow poison. It is used chiefly as a bland article of food in the sick-room in the form of light pudding or other desserts, but may be combined with other starch foods and made into bread. There are several varieties, the best of Avhich come from Bermuda and Jamaica. Corn starch is employed frequently as a fair substitute. Arrowroot starch is shown in Plate XI., Fig. 2. 3. Fatty Seeds (Nuts). Nuts are rich in fat and })roteids, but contain no starch. They are of high nutritive value, but on account of their richness in fat they are not easily digested, even when reduced to a finely divided state. ALMONDS. In the countries where they are produced, the almond is eaten both in the green and diy conditions. The ripe kernel has a skin, with a bitter disagreeable taslc. When this is removed by soaking for a time in warm water, the almond is known as " blanched." There are two varieties of almond, the sweet and the bitter, both of which contain more than 50 ])er cent, of oil, about half as much pro- teid material, gum, sugar, and crude fiber. Both contain emulsin, a substance Avhich, in the presence of water, acts upon the glucoside amygdalin, present only in the bitter variety, to form hydrocyanic acid, glucose, and benzoic aldehyde. On account of this reactiou, the bitter almond is not always safe, and fatal results have occurred from its ingestion. A¥hen almonds are baked, they are made more brittle, and are re- duced more easily to a powder. COCOANUTS. The fleshy white kernel of the cocoanut contains about 70 per cent, of fat. The milky interior is chiefly water, but contains nearly 7 per cent, of sugar. WALNUTS. All of the trees of the genus Juglcms yield nuts classed as walnuts. The different varieties, though varying in outward appearance and in taste, have practically the same composition. They contaiu about GO per cent, of fat, about 16 per cent, of proteids, and about 7 per ceut. of sugar and gum. The hazel nut, which belongs to the oak family has about the same composition. OLIVE OIL. 153 PEANUTS. The peanut, known also as ground nut and goober, is less rich in fat, but richer in proteids than other nuts. It contains about 45 per cent, of the former and about 30 per cent, of the latter. CHESTNUTS. The chestnut is not of this class, but for convenience will be con- sidered here rather than with the farinaceous seeds, in which class it properly belongs. It contains but little fat and proteids, about 15 per cent, of sugar, about 25 per cent, of starch, and about 50 per cent, of moisture. It is very indigestible in the raw state, and even when cooked is very trying to the digestion of those with weak stomachs. It is used very extensively as a food by the French, Spanish, and Ital- ian peasantry in various cooked forms, and largely in the form of bread. 4. Vegetable Fats. The vegetable fats include the oils derived from the olive, cotton- seed, peanut, and other seeds. They are used in the preparation of salads and for frying. The most important are the two first mentioned. OLIVE OIL. Olive oil is a bland fixed oil derived from the fruit of the many varieties of the olive tree. It is known bv various names which desis:- nate the grade, but is sold for the most jDart as virgin oil, ^vhich is the choicest grade of all and not extensively marketed. Virgin oil is made from the choicest olives, about three-fourths ripe, which are bruised only slightly in the mill, so that only the olive pulp, and not the stone, is crushed. The crushed mass is gathered in a heap, and the oil is allowed to drain away without pressure or other influence of anv kind. The product has a greenish tint and a far more delicate taste than that made in the manner to be described. In the manufacture of the grades ordmarily seen in the market, the olives, both pulp and stones, are ground into an oily paste, which is packed uito bags made of woven grass. These are placed in piles and subjected to pressure. As the oil drains away, boiling water is applied to the bags to keep up the flow, and that which is thus obtained con- stitutes the lower grade. Sometimes, the pressed pulp is thro^^m into water and separated from the broken kernels, which sink to the bottom. The pulp is then gathered up and pressed again. On account of the cost of pure olive oil, adulteration with other cheaper oils is practised very extensively. The principal adulterant is cotton-seed oil, which is exported from this country in large quantities for this and other purposes. INIuch of the oil sold in this country as olive oil is cotton-seed oil put up in the cheapest kinds of bottles, adorned with gaudy labels bearing inscriptions often not remarkable for accuracy in the use of the French language. The author has seen, for 154 FOODS. example, labels which indicated that the contents of the bottles had been " virginated." Adulteration of olive oil to only a slight extent with the cheaper oils is by no means easy of detection, but when the fraud is fairly ex- tensive it may be shown by chemical tests and by the use of the re- fractometer, the retractive index of olive oil bemg less than that of the cheaper substitutes. The iodine number and saponification equiv- alent of olive oil are both less than those of its adulterants. The be- havior of olive oil in contact with nitric acid t)r with alcoholic solution of nitrate of silver is markedly dilferent from that of the cheaper oils. Thus, equal volumes of strong nitric acid and olive oil, niLxed together and agitated in a flask, give a product which has either a greenish tinge or at most one inclining to orange, and no marked change is perceptible on standing for five or ten minutes ; whereas cotton-seed oil similarly treated yields almost immediately a reddish color, Avhich short S' darkens and becomes dark brown or almost black. xVgain, if 12 cc. of a suspected sample are mixed in a test tube with 5 cc. of a 2.5 per cent, solution of nitrate of silver in 95 per cent, alcohol, and placed in a beaker of boiling water, the resulting change of color gives indications as follows : if olive oil, the color is greenish ; if cotton-seed oil, it becomes black ; if sesame oil, it is dark reddish- brown ; if peanut oil, it is at first reddish brown, then greenish and turbid ; if poppy oil, it is greenish yellow. For further details of chemical tests, the reader is referred to the standard works on the adulteration of foods. COTTON-SEED OIL. This very important and cheap vegetable fat is a perfectly whole- some and desirable article of food. It is much used under its own name as a substitute for lard and olive oil for frying, and in ])lace of the latter as an ingredient of dressings for salads. It lacks the fine flavor of olive oil, but its substitution in dressings can be detected only by the educated palate. From a hygienic standpoint, there is abso- lutely no objection to its use in the prc]iaration of foods. The same may be said of the other cheap vegetal)le oils. 5. Tubers and Roots. In the cooking of tubers, roots, and other vegetables, the albumins and globulins are coagulated, the fibrous matters in the cell walls are softened and ruptured, the starch granules swell and burst, the starch itself becomes somewhat chanared in character, and the whole mass is made more digestible. When boiling is the process employed, part of the mineral matter and more or less of the other soluble substances, including certain protcid material, are extracted and lost. POTATOES. The potato is the most important member of this group. It was intro- duced into Spain from Peru about the middle of the sixteenth century, PLATE XII. >!re})aring them for the market. The choicest varieties, for example, are those which include only the terminal leaves, and the poorest those made up of the largest and coarsest leaves from the lower end of the twig. TEA. 169 Tea is classed commonly as greeu or black. Both kinds come from the same shrub, but are different in point of age, and are cured in dif- ferent ways. Green tea is made from young leaves, which are roasted quickly shortly after being gathered, and then rolled and again roasted. Black tea is from older leaves, which are allowed to wilt, and then are gathered into heaps and left without farther manipulation for about a half day, during which time they undergo a fermentative process and change color. Next, they are rolled by hand and then heated, and these processes may be repeated several times alternately. Finally^ they are dried slowly over burning charcoal. The composition of tea is very variable, and it is impossible to give figures which may be accepted as indicating the approximate constitu- tion of a typical specimen. K5nig has collected 16 analyses, which give the following averages : Moisture 11.49 Nitrogenous matters 21.22 Tlieine 1.35 Volatile oil 0.67 Fat, resin, etc 3.62 Gum, dextrin, etc 7.13 Tannin 12.36 Other extractives 16.75 Fiber 20.30 Ash 5.11 100.00 But it should be said that the variations in the amounts of individual constituents of these 16 specimens are very wide : for instance, water, 4.59 to 16.06 ; theine, 0.40 to 4.94 ; tannin, 4.10 to 20.88 ; fiber, 15.11 to 25.06. Dragendorff found, in 23 specimens, from 1.36 to 3.09 per cent, of theine, 7.10 to 12.66 of moisture, and from 24.80 to 44.50 per cent, of total soluble constituents. The ash of pure tea is fairly constant in amount, and almost never reaches as high as 7 per cent. ; usually, between 5 and 6 per cent. Tea should be used only in the form of an infusion, made by pouring boiling water upon the requisite amount of leaves, and allowing it to stand a short while to " draw." It is used not uncommonly in the form of a decoction ; that is, by boiling. This process is objectionable in two ways : first, the delicate aroma is lost by the expulsion of the very volatile essential oil ; and second, the leaves are made to yield all their tannin and other extractives, which tend to bring about, sooner or later, derangement of the digestive function and a catarrhal condition of the stomach. The finest and most delicate portion of an infusion is that which is poured off within three or four minutes, for in this will be found a maximum of flavor with a minimum of bitterness and astringency. The excellence of an infusion is influenced considerably by the character of the water, which, if very hard, is slow in extracting the desirable soluble constituents, while, if very soft, it extracts not only these, but far too rapidly the less desirable principles. When properly made, tea in moderation is a wholesome, agreeable, 170 FOODS. and refreshing stimulant beverage, particularly grateful in conditions of mental or physical weariness. Used in excess, it exerts a harmful influence upon the nervous system, and in a too strong form iujures the digestive tract and function. The abuse of tea as a beverage leads, according to Bullard,' to ring- ing in the ears, tremor, nervousness, headache, neuralgia, hysteria, irregularity of the heart, dyspnoea, dyspepsia, and constipation. Dr. Hayes, the Arctic explorer, has testified to the value of tea and coifee in enabling men to endure cold and hardship of all sorts, tea being especially soothing at the end of a hard day's work. AVhile tea by itself can hardly be looked upon as an article afford- ing any important amount of nutriment, as commonly consumed it serves as a vehicle for other substances, as sugar, milk, and cream, having high nutritive value. Adulteration of Tea. — It is commonly stated and generally be- lieved that tea is adulterated extensively with other kinds of leaves, including those of the beech, sloe, willow, and hawthorn ; but at the present time, it is extremely improbable that such adulterants ever are mixed with tea known to be intended for export to this country. AVhatever the conditions may have been prior to the enactment of the national law governing tea importation, the fact now is that our tea sup])ly is practically free from this form of adulteration. The detec- tion of spurious tea leaves would be an easy matter, since the genuine have a very characteristic appearance which can hardly be confused with that of any of the possible substitutes ; and even when broken into small bits, the characteristic differences in venation and serration, and in the stomata as well, are plainly discernible. ]More probable forms of adulteration include the admixture of wholly or partially exhausted leaves ; the addition of astringent mat- ters, such as catechu, to lend color and apjiarent strength to the infusion ; the ]iresence of foreign mineral matter ; and the ])ractice of " facing." The presence of any large proportion of exhausted leaves can be detected by the low amount of total soluble extract and by the small amount of soluble ash, which should not be less than 3 per cent, of the weight of the leaves. The presence of important amounts (tf accidental or added mineral matters is shown in the total ash, which in a genuine specimen rarely amounts to 7 and never to 8 per cent. The substances most often found are sand and soapstone ; the first named is found some- times in amounts exceeding 25 per cent. Catechu is applied occasionally to exhausted tea leaves with the aid of solutions of gummy matters, for the purpose of adding astringeucy and color to the infusion. Teas so treated have but little, if any, of the true aroma, and their infusions yield a sediment in which the i)ar- ticles of catechu can readily be seen. The object of " facing " is to make the product appear to l)e of greater value, and the practice is, therefore, pr(^]>erly speaking, one M'hich comes within the definition of fraudulent adulteration. Damaged ' Boston Medical and Surgical Journal, April 8, 1886, and September 8, 1887. COFFEE. 171 or otherwise inferior leaves are treated with Prussian blue, plumbago, indigo, and other substances, and the small amount which adheres im- proves their color and general appearance. This amount is too small to be of any sanitary significance. The presence of facing materials may be detected by the use of the microscope and by chemical analysis. COFFEE. Coffee is the seeds of the Coffea Arahica, dried and deprived of their fleshy covering. The fruit is a small pulpy berry containing, usually, two seeds. The tree is said to have originated in Abyssinia, where, however, in the seventeenth century there were few, if any, specimens, and to have been introduced into Arabia in the fifteenth century. It is now grown very extensively in Brazil, Java, Peru, Ceylon, West Indies, and other hot countries. The first European to mention it was Prosper Alpinus, of Padua, who included it in an account of Egyptian plants, published in 1592. The first work devoted wholly to coffee was a small Latin treatise. Be salubemma potione cahue, by Faustus Nairo, Rome, 1671. Coffee was first sold in London by a Levantine, in 1650, and some years afterward was introduced into France. The first whole cargo introduced into this country arrived in 1809, but coffee houses were licensed in Massachusetts as early as 1715. The world's production of coffee for the year ended June 30, 1900, was estimated at almost 900,000 tons.^ This country alone consumes more than the whole of Europe: in 1897 we consumed 318,170 tons against 305,150. The total consumption by Germany was 136,390; by France, 77,310; by England, 12,420; and by Italy, 12,500 tons. As is the case with tea, coffee must undergo a process of roasting before it is fit for use, although it is said that the Arabians and other Eastern peoples make a decoction of the raw article and swallow the grounds as well as the liquid. The roasting is conducted at about 200° C. until the natural color, which is greenish, grayish, or drab, is changed to a rich dark brown. During the process, certain volatile aromatic principles are developed, the alkaloid caffeine is dissociated from its union with tannin, the moisture is very largely expelled, the sugar is caramelized, gases are formed (largely carbonic dioxide) which cause the berry to swell, and much rupturing of the cell layers occurs. The berry thus loses in weight and gains in bulk. The process must be conducted carefully, else the quality will not be what is desired, since if the roasting is not pushed sufficiently far, there will be insufficient development of aroma ; and if it is carried too far, the volatile matters are expelled and the product acquires an unpleasant taste. On account of the volatile nature of the aromatic principles developed, coffee should be roasted only as the demands of commerce make it necessary. On long keeping, except in hermetically sealed containers, it undergoes extensive deterioration. For the same reason, the roasted berries should be ground only as needed. ^ Consular Reports, Vol. LX., p. 258. J 72 FOODS. Coffee contains less caffeine (theiue) than is found in tea ; thus, Drag- endorff found the amount in 25 samples to vaiy betsveen 0.64 and 2.21 per cent., Avhereas in about the same number (23) of samples of tea, the range Avas 1.36 to 3.09. It contains considerable amounts of fat, generally over 12 per cent., about the same amount of nitrogenous matters, small, quite unimportant amounts of sugars, gummy matters, and other substances, and about 40 per cent, of fiber. Coffee is used in infusion and as a decoction. Like tea, it loses its; pleasant aroma when boiled, but its decoction is less bitter and astrin- gent than that of tea. In order to enjoy both the fragrance of an in- fusion and the strength and body of a decoction, it is not an uncommon practice to make first the one and pour it off, and then, with a fresh portion of water, to boil the grounds for a few minutes, and then to mix the two liquids together. Coffee acts as a decided stimulant to the nervous system, enables one better to perform arduous work, and diminishes the sense of fatigue. In small amounts, it increases the force and frequency of the pulse, but taken in excessive quantities, it causes ])alpitation and intermission, besides general nen'ousness and derangement of digestion. It has a marked inhibitor^' influence on gastric digestion, and is more oppressive to the stomach than tea and, hence, should be used with caution by dys- peptic-^. AVith some persons it stimulates peristalsis, and thus acts as a gentle cathartic. It increases the secretions of the skin and kidneys. Coffee is adulterated ver\- extensively Avith a variety of substances of widely different nature, including chicory, dandelion, and other roots, roasted cereals and legumes, sawdust, date stones, red slate, acorns, and other cheap articles. It is not alone in the ground form that it is falsified, for even the beans are imitated with mixtures of flour and other materials, moulded to the correct shajie and carefully roasted and colored. The detection of adulterants in coffee requires l)Ut little time. Of great assistance is the fact that coffee contains absolutely no starch, while most of the commoner adulterants contain it in alnnidance. Therefore, if a specimen under examination is boiled and filtered, and the filtrate gives a dirt\' blue reaction with test-solution of iodine, one may be sure that adulteration ha> lieen ]>ractised. But not all of" the adulterants are starchy in their nature and, therefore, other examination is necessary. ^Microscopical examination will detect not only the starchy, but the non-starchy matters as well. Under the microscope, ground coffee has a characteristic aj)pearance which cannot be mistaken for any- thing else. Chicory and other roots, date stones, and all other berries and seeds have their own characteristics. For the mere determination of the question of juirity, only a knowledge of the microscopical ap- pearance of coffee itself is required, and this is acquired easily and quickly by direct study. For the identification of the adulterants present, one necessarily should be familiar with the ajipearance of all of the substances used. Chicory is the root of the Cichorinm infybus, a perennial herb, grow- COCOA. 173 ing wild and extensively cultivated in this country and in Europe. The roots are cleaned, cut into pieces, dried in kilns, roasted in iron cylin- ders, and ground into a coarse powder. Like coifee, chicory when roasted contains a volatile principle and a bitter. It is used both as an adulterant and as a substitute for coffee. Mixed with coffee, it lends both color and flavor to the infusion, and by many is regarded as a ■desirable addition. It itself is subject to adulteration by cheaper roots, such as mangel wurzel and dandelion. Coffee and chicory behave very differently when thrown into cold w^ater : the former floats and retains its firm consistence, while the latter absorbs water very quickly and sinks, and in its descent leaves streaks of color. Coffee wliich has been roasted too much will, how- ever, sometimes sink, and chicory which has been treated with fatty substances will float. Mixtures of the two can often be detected by the difference in resistance when placed between the teeth. The par- ticles of coffee are much harder than those of chicory, which yield very readily to pressure and also have a sweetish taste. Inferior and damaged raw coffees not infrequently are colored and faced, in order that they may be improved in appearance or be made to imitate better grades. The facing agents used are mixtures contain- ing variable amounts of ultramarine, indigo, clay, gypsum, chromate of lead, and coal dust. According to G. Wirtz,^ inferior grades of coffee are treated largely at Antwerp, Rotterdam, Hamburg, Bremen, and elsewhere, by washing, coloring, and finally drying by centrifugation with sawdust, the result I)eiug a fine white product of an apparently greater value. Package coffees sold under various names, such as " French Break- fast Coffee," " A'^ienna Coffee," and " Eureka Breakfast Coffee," are rarely anything more than roasted and ground cereals and peas. It is to be said, however, that their character usually is indicated in the directions for use printed on the labels, which commonly begin by advising the use of " a thuxl more than you would use of genuine coffee." Microscopical examination and the iodine test will reveal their composition very quickly. COCOA. Cocoa, a corruption of Cacao, and in no way related to the cocoanut, is derived from the seeds of the Theobroma cacao, a native of tropical America. It is estimated that the annual production of the seeds amounts to about 150,000,000 pounds, more than a fifth of which is exported by Ecuador alone. ]N"early a fifth of the annual crop is con- sumed within the United States. The fruit of the cocoa tree is a pod, about a foot long and half as wide, filled with " beans," or " chocolate nuts," about as large as almonds, imbedded in five rows of from four to ten each in a pulpy matrix. When ripe, the pods are gathered and collected into heaps, and left for a day or longer ; then they are cut open and deprived of ' Zeitschrift fur Untersuchung der Xahrungs- und Genussmittel, 1898, p. 248. 174 FOODS. the seeds, which are allowed to undergo a ]5rocess of fermentation in earthen vessels or in holes in the ground. This process, which must be regulated very carefully, has fu' its ol)ject the I'cmoval of an acrid, bitter taste and consequent improvement in flavor. Sometimes, the seeds are dried in the sun as soon as removed, but the product is then of much less value ; sometimes, the entire pod is buried until the pulp becomes rotten and softened. AMien the fermentation process is com- ])lctcd, the seeds are dried carefully in the suu, and then become hard, brittle, and reddish or reddish brown in color. In the preparation of cocoa for the market, the seeds first are cleaned and carefully roasted. As is the case with coffee, the roasting must be carried to a certain point to insure the development of the desired flavor, but not so far beyond as to impair it. During this process, the thin husks of the seeds become more detachable, and before the next operation they are removed. Then the seeds are crushed lightly and freed from their hardened germs, and in this form are knoAvn as " nibs." These arc gronnd in a special form of mill into a paste (" flake cocoa "), which is moulded into cakes and allowed to harden. In this form, the product is known as plain chocolate. The sweet and flavored choco- lates are made with the addition of sugar, vanilla beans, cinnamon and other s})ices. Inferior vanilla chocolate is made with artificial vanillin and coumarin, in ])lacc of the far more expensive and better flavored vanilla bean. For the preparation of powdered cocoa, it is necessaiy to remove a part of the oil, which, when present in its normal amount, favors cak- ing. This removal is accom])lished by hydranlic pressure, and the paste is then ])assed through sieves of exceedingly fine mesh. The so-called soluble cocoas are prepared with sugar and starches, particularly arroMroot, but the cocoa itself is not soluble in water. The apparent solubility is due to the fineness of the powder and to the in- crease in the specific gravity of the liquid due to the sugar in solution, both these conditions favoring prolonged suspension without sedimen- tation. Some of the Dutch soluble cocoas are treated with alkalies, for the removal of the crude fiber and for their effect u]ion the coloring matters. These cocoas thereby lose j^art of their natural flavor, but the loss is made up somewhat by the addition of fragrant foreign matter. Cocoa was introduced into Europe by the Spaniards after their in- vasion of Mexico under Cortez, in 1519. It was not known in England until 1657, when it was sold first in London by a Frenchman. In this country, it first was prepared and sold at Danvers, Massachusetts, in 1771, from raw material brought from the West Indies l)y the fisher- men of Gloucester. Unlike tea and coffee, which in themselves can hardly be regarded as adding any nutriment to the diet, cocoa is an exceedingly valuable food, which possesses the advantage of much nutriment in small bulk, and hence is particularly suited to the needs of those engaged in expeditions removed from civilized centers. It makes a wholesome, agreeable, stinuilaut beveratie, and is eaten in the form of chocolate, BEER. 175 and as an addition to cakes, puddings, and other compounds. The cocoa nibs and plain chocolate contain about 50 per cent, of a whitish solid fat of agreeable taste and smell, commonly known as cocoa, butter. It contains variable amounts of the alkaloid theobromine (dimethylxanthine), which is related very closely to caffeine and theine (trimethylxanthine), and has nearly the same physiological action, although somewhat less stimulant and rather more diuretic. The amount is said to average about 1.50 per cent. Cocoa is rich in nitro- genous matter, contains more than 10 per cent, of a starch with small rouud granules, and about 3.50 per cent, of ash, which is largely phos- phate of potassium. Sixteen analyses of the kernels, compiled by Konig, give the follow- ing averages : Water 3.63 Proteids 13.49 Fat 49.32 Starch 13.25 Exti-actives 13.18 Fiber 3.65 Ash _3^8 100.00 The husks, commonly known as " shells," are used in the prepara- tion of a cheap and wholesome beverage. They contain little fat, but are about equal to cocoa in nitrogenous matter, and contain more than 40 per cent, of nitrogen-free extractives. Cocoa and chocolate are subject to extensive adulteration with sub- stances having much less commercial value, though j)erhaps equally nutritious. Among those used, are starches of various kinds, as wheat, rye, potato, arrowroot, and rice, sugar, vegetable oils, mutton tallow and other fats, Venetian red, clay, and brick dust. Various flavorings are employed, such as vanillin, coumarin, clove, mace, cardamom, and nutmeg ; but unless these are used under the name of vanilla or of other flavorings than themselves, they cannot be regarded as adul- terations. Fermented Alcoholic Beverages. BEER. Beer is the generic term which includes all fermented drinks made from malt — lager beer, ale, porter, and stout. As commonly under- stood, beer is an infusion of malted barley, flavored with hops and fermented with yeast ; but on account of the fact that wholesome sub- stitutes for malt and hops may be employed in its manufacture, it is defined also as a " fermented saccharine infusion to which some whole- some bitter has been added." In this country, the term beer is restricted commonly to the product generally known as lager beer. Porter is a beer with a high percentage of alcohol, and is made from malt dried at a high temperatui-e. Stout contains less alcohol and hops, but more malt extract. Ale is a pale beer containing more hop extract 176 FOODS. and less malt extract than porter or stout, and brewed by " top fermen- tation." Beer was made by the Egyptians many centuries before the Christian era. It is related that, for public reasons, the suppression of beer-shops was attempted by their government more than forty centuries ago. The art of brewing was taught by them to the ancient Greeks and Romans ; thus, beer was a common drink in Greece prior to 700 B. c, and was One of the principal beverages of the soldiers of Caesar. In the time of Tacitus, it was in common use in Germany ; and Pliny speaks of its use in Spain. The ancient Britons, at the time of the Roman conquest, made it from barley. At the time of the Xorman conquest, the words beer and ale meant in England the same thing : a drink made of malt without hops. Later, the word beer fell into disuse ; but in the fif- teenth and sixteenth centuries, after the introduction by the Flemish of beer made with ho]is, the term was revived, and then meant hopped ale. The use of hops was forbidden in 1530 by Henry VIII., who regarded them as an adulterant, and in the first year of the reign of Richard III., the authorities of London laid a fine of 6s 8d on every barrel (»f beer containing them. Later, this was reduced (me-half. The prejudice against the use of hops in brewing is expressed by one of the earliest English writers on dietetics, Andrew Boorde,^ Avho says : "Ale is made of nialte and water; and they the which do ])ut any other thynge to ale than is reherscd, exce])t yest, barme, or godesgood, doth soiysticat theyr ale. Ale for an Englysshe man is a natural drynke. Ale must hauc tliese property es : it must be fresshe and cleare, it muste not be ro])y nor smoky, nor it must hauc no weft nor tayle. Ale shuld not be drunk under v. dayes olde. Newe ale is vnholsome for all men. And sowre ale, and dcade ale the which doth stande a tylt, is good for no man. Barley malte maketh better ale than oten malte or any other corne doth : it doth engendre grose humoures ; but yette it maketh a man stronge. Bere is made of malte, of hoppes, and water : it is the naturall drynk for a Dutche man. And nowe of late dayes it is moche vsed in Englande to the detryment of many Englysshe men ; specyally it kylleth them the Avhich be troubled with the colycke, and the stone, & the strangulion ; for the drynke is a colde drynke; yet it doth make a man fat, and inflate the bely, as it doth appere by the Dutche mens faces & belyes. If the bere be well serued, and be fyned, & not newe, it doth qualyfy the heat of the lyuer." The ancient Germans made beer from all kinds of grains, and for flavoring used oak bark, sage, and leaves of the laurel, ash, and tama- risk. IIo]is were used more or less from the ninth century, and came into general use in the eleventh. Beer being the common drink of most European peoples before the establishment of colonics in America, it followed naturally that the early settlers of this country brought the art of brewing with them. In 1629, the cultivation of ho))s had been carried on for some time in ' A C()n)])en(ly<)iits Kegyment, or A Dyetarv of Helth, maile in Mountpylier, ooin- pyled by Andrewe Boorde of Physyche Doctour, London, 1542. BEER. 177 New Amsterdam, and hop roots were sent for from England by the authorities of Massachusetts. In nearly all the colonies, the brewing of beer was regarded as quite as essential an accomplishment of women as the abihty to make good bread. The first law regulating the sale of alcoholic beverages in Massa- chusetts was made in 1633 ; it prescribed that no person should sell wine or spirits Avithout a permit, but made no reference to beer. In the following year, it was ordered that no one should charge more than a penny for a quart of beer, and in 1637, that no inn-keeper or vic- tualler should sell any intoxicating drink but beer ; and this they were prohibited from brewing themselves, but must obtain from a licensed brewer. In the following year, owing to the fact that the only one of this class was unable to meet the demand, they A\'ere allowed to conduct the process themselves. In 1649, it was ordered further that every inn-keeper and victualler should keep always on hand a supply of good, wholesome beer. In 1651, the court undertook to stimulate the pro- duction of a better grade of beer in the belief that thereby the growing tendency to the use of wine and spirits and the increasing habit of drunkenness would be checked, and permission was granted to charge one, two, and three pence per quart, according to the amount of malt used per barrel. A duty of a shilling per bushel of imported malt, imposed in 1654, called forth a protest from Boston merchants, on account of the very great importance of beer as a beverage of the people. In the following year, in order to promote home production of malt, importation Avas prohibited, but this order was repealed in 1660. In 1667, the use of molasses as an adulterant of beer was punishable by a fine of five pounds. Similar laws relating to beer were passed from time to time by the authorities of all the original colonies. Process of Manufacture of Beer. — The first step in the brewing •of beer is the preparation of the malt. The barley first is steeped in water for several days, and then is removed and arranged in heaps, which, after a time, are spread out and turned repeatedly until germi- nation has proceeded to the requisite extent. Xext it is dried in kilns at a temperature below or about 90° F., and then is heated to from 125° to 180°, according to the color desired. This process develops flavor, completely checks germination, and determines the commercial character of the product. The steeping of the malt is done best in water containing considerable of the mineral salts that cause hardness ; a soft water exerts too much solvent action on the proteid matters, which, soon after extraction, are likely to undergo decomposition. During the progress of germination, the ferment •diastase is developed, and proceeds to convert the starch into dextrin and maltose. After the germs and rootlets have been removed by proper screening and sifting, the malt is crushed, and then an infusion, the "wort," is made with water at about 160° F. This is drawn off from the exhausted malt, and then boiled for an hour or two with hops, which, besides giving a characteristic bitter flavor, assist in clarification 178 FOODS. by the action of their contained tannin on some of the proteid matters. Then the boiled bitter wort is cooled rapidly, run into vats, mixed with yeast, and allowed to ferment for several days, during which time alcohol and carbonic acid are formed from the maltose. The natm'e of the product is influenced very largely by the purity of the yeast and by the method of fermentation followed. Top fermentation is carried on rapidly, and at a compai'atively high temperature, the yeast growing at the surface ; in bottom, or sedimentary, fermentation, the yeast grows at the bottom, the process is slower, and is carried on at a lower temperature. The chief advantage of the employment of yeast which grows at a low temperature is that other, perhaps undesirable, growths may be unable to })roceed. Whatever the process of fermenta- tion followed, not all the sugar should be allowed to be converted, for then the flavor would be not what it should, and the keeping qualities would be impaired. On the completion of fermentation, the beer is separated from the yeast and transferreal preservative used in beer is salicylic acid ; next in im])ortancc is fluoride of sodium, which, however, is not used to any considerable extent as yet in this country. Salicylic Acid. — The ordinary method of extracting by means of ether and testing the residue left on evaporation of the latter with ferric chloride, cannot be used in the examination of beer, since kiln- dried malt contains a principle which gives a reaction identical with that of salicylic acid. The following method, devised bySpica, is, how^ever, satisfactory and reliable : Acidify 1 00 cc. with sulphuric acid, extract with ether, allow the separated ether to evaporate sponta- neously, and warm the residue gently with a drop of strong nitric acid, W'hereby, if salicylic acid is present, i)icric acid is formed. The addi- tion of a few dro])s of ammonia or of sodium hydrate produces the corresponding |)icrate with its bright-yellow color, -which may be imparted to a woollen thread immersed in the liquid. Fluorides. — Several methods are recommended, and among them the following : Method of Hefelmann and Manx. — Expel the carbonic acid WIXES. 189 from 500 cc. of beer, and then add 1 ec. of a solution containing 5 per cent, each of calcium and barium chlorides, and follow it with 0.5 cc. of 20 per cent, acetic acid and 50 cc. of 90 per cent, alcohol. Let stand twenty-four hours and filter. Dry the filter and precij^itate col- lected thereon without w^ashing, and transfer to a platinum crucible. Add strong sulphuric acid, and cover the crucible with a waxed watch- glass with some lines scratched through the wax coating, then heat at 100° C. for two hours, and obser\^e the effect on the exposed glass. This method is said to be of sufficient delicacy to detect the presence of 7 milligrams in a liter. Brand's Method. — To 100 cc. of beer made slightly alkaline with ammonium carbonate and heated, add 2 or 3 cc. of a 10 per cent, so- lution of calcium chloride. Boil for a few minutes, filter, and dry the filter and contents. Then proceed as in the method just described. In either process, it is best to place a lump of ice in the conca'sdty of the watch-glass to keep the latter cool ; the water should be removed from time to time by means of a pipette so that it may not overflow. Other Determinations. — Of minor interest are the determinations of acidity and ash. Total Acidity. — To 10 cc. of beer freed from carbonic acid by shak- ing, add a few drops of neutral litmus solution, and then add decinor- mal sodium hydrate until the end reaction is observed. Express the results in parts of acetic acid. One cc. of decinormal sodium hydrate equals 0.006 gram of acetic acid. Fixed and Volatile Acidity. — Concentrate 10 cc. of beer to a third of its bulk by evaporation, add water up to the original volume, and proceed as above. The difference in results is due to the acetic acid which has been driven off. The fixed acidity is due chiefly to lactic acid, and, if desired, may be so expressed. One cc. of the decinormal solution LS equivalent to 0.009 gram of lactic acid. The other acids present mclude succinic, malic, and tannic. Ash. — The residue obtained in the direct determination of the ex- tract may be utilized for the estimation of the ash. It should be ignited very cautiously and at as low a temperature as possible until the ash becomes white. WINES. Properly speaking, wine is the fermented juice of grapes, though the term is applied also to other products of fermentation of saccharine liquids and fruit juices. It has been in use as a drink from the very earliest periods of civilization. At the present time, wines are pro- duced in infinite variety and of many qualities. The character and properties depend upon a great number of factors, including the variety of the grape, the nature of the soil upon which the vine is cultivated, the climate in general, and the state of the weather in particular when the grapes are ripening, the degree of ripeness when gathered, the method followed in the preparation of the must, and the care with which the other steps in the making of the final product are conducted. 190 FOODS. Of very great influence is the extent to which the seeds, skins, and stems of the fruit are allowed to he acted upon. The seeds yield con- siderable amounts of tannic acid, and the skins lend color, flavor, and to some extent astringeucy. The most important constituent of the juice of the grape is the sugar, and this is present in greatest alnm- dance when the fruit is fully ripe. In the making of wine, the first step is the preparation of the must. The grapes, with or without preliminary careful examination and sort- ing, usually without, are crtLshed by machineiy or by the naked feet of men, so that the juice is set free. Sometimes, the stems are first carefully eliminated, and particularly good individual grapes are cut out and set aside for special use. In the crushing of the fruit, the method of treading has in its favor the fact that the seeds are not thereby affected, and so do not give up so much of their astringent principle. If a white wine is to be made, the must is freed at once from the skins and stalks ; but if the product is to be red, these are retained during the process of fermentation. The juice of both the white and the black varieties of grapes is practically without color ; but when the dark skins are left in contact with the fermenting mass, the alcohol formed extracts the yellow and blue coloring matters, which become red under the action of the free acids formed at the same time. The constituents of the must are water, sugar, proteid matters, gummy substances, pectous matter, organic acids and their salts, and mineral matters. The must, with or without the skins and seeds, is fermented in vats of wood, marl)le, or stone, the process starting very quickly, lieing in- duced by organisms which grow on the skin itself. The temperature at Avhich this Ls allowed to proceed exerts an important influence in determining the character of the wine: conductefl between 5° and 15° C, the process is comparatively slow and the aroma of the wine is rich • while at higher tcm]ieratnres, the rate is more rapid and the bouquet is less marked. The tennination of the process is made evident by cessation of the evolution of carbonic acid, the diminution of specific gravity, and the sinking of matters ^\•hich before had formed part of the scum. Whether all of the sugar is used uj), depends somewhat upon the amount of proteid nutrient material for the growth of the organisms by which the conversion is carried on. If this is exhausted first, there will be a residue of sugar, and the product will be coiTcspondingly sweet ; if there is an abundance of proteid matter, the sugar will be the first to be exhausted, and the wine will be ''dry." It is sometimes necessary- to add nitrogenous matter, such as egg albumin or gelatin, in order to keep the process from ceasing too early. As the ]iercentage of alcohol in the fermenting must rises, the bitar- trate of potassium present is depcsited gradually, owing to its insolu- bility in alcohol. The deposit is known commercially as argol, and is the source of cream of tartar. When the first fermentation is completed, the alcoholic liquid is drawn WINES. 191 off into casks, in which it is kept for a number of months, the vessels being kept constantly filled. It now undergoes a second slow fermen- tation, which brings about changes which are not understood excepting in their gross result, which is the production of the "bouquet" or flavor. In this second process, there occur a farther deposition of argol and an oxidation of aldehyde to acetic acid. The bouquet is due to a combination of ethers, the chief of which is oenanthic ether, sup- posed to be produced from the alcohol through the agency of the organic acids normally present. The wine next is racked off into other casks, and in some cases it is necessary to do this several times. Sometimes, the appearance of the wine is such that "fining" is necessary. This consists in the addition of egg albumin, isinglass, or other gelatinous matter, v/hich in its descent attracts and enmeshes the fine particles of matter which not only pre- vent brilliancy, but later on may impair the keeping quality of the wine. Classification of Wines. — Wines are classified variously according to color, strength, sweetness, and content of carbonic acid. Accord- ing to color, they are classed as red or white, the latter term applying not only to the very light, almost colorless kinds, but also to those having a decided yellowish or even yellowish-brown color, such as is possessed by " white port." The red wines include those generally known as Clarets and Burgundies, though both these kinds exist in the white forms. Tlie white wines include the white Clarets commonly designated as Sauternes, white Burgundies of which Chablis is a type, the Rhine and Moselle wines, and others. According to strength, wines are classed as natural and fortified. The natural wines contain of alcohol only that which is formed in the process of natural fermentation ; the fortified wines, such as Sherry, Port, and Madeira, contain, besides, a considerable amount in the form of added spirits. According to their content of sugar, wines are classed as sweet or dry. Some of the sweet wines contain added sugar and that which has escaped the action of the yeast plant. In the dry wines, all or nearly all of the sugar has been converted into alcohol. Not all of the sugar, however, in any wine is converted into alcohol and carbonic acid, small amounts going to form glycerin and succinic acid. According to their content of carbonic acid, wines are classed as still or sparkling (effervescent). The natural wines contain practically no carbonic acid ; the sparkling, or effervescent, wines, as Champagne and sparkling Moselle, are in a sense artificial in that they are subjected to a process of fermentation in the bottle, sugar being added for the pur- pose. They are flavored also with liqueurs. Composition of Wines. — Alcohol. — The most important constitu- ent, the active principle, of wine is ethylic alcohol. The higher alcohols, propylic, butylic, and amylic, are always present in traces. The amount of alcohol is variable, ranging in natural wines from 6 to 14 per cent, by weight, but ordinarily present between the limits of 192 FOODS. 9 and 12 per cent. lu fortified wines, the amount ranges from 12 to about 22 per cent., but is usually about 17 per cent. Sugar. — While the amount of sugar in the original nuist ranges be- tween 12 and 33 per cent., in the natural finished product it is as a rule quite low, ordinarily considerably under 0.5 per cent., and often none at all. The sweet Tokays contain exceedingly variable amounts, ranging from 3 to 20 per cent.. Ports and Madeiras about 4, and Sherries some^^'hat less ; but American Ports, Sherries, and ^Madeiras are commonly fairly rich in sugar. Domestic Champagnes, also, con- tain notable amounts, but those of foreign origin, even those ordinarily classed as sweet, contain but small amounts, the impression of sweet- ness being largely due to the flavorings of the liqueurs added. Four specimens analyzed by the author, one of which (Xo. 4) is well known as an extra sweet wine, yielded the following results : Brand. Sugar. Extract. Alcohol by weight. 1. Brut Imperial (Moet & Chandon) 1.35 1.52 1.56 4.76 3.27 3.21 11.15 10.38 3. Dry Imperial (Moet & Chandon) 4. Wiiite Seal (Meet & Chandon) 3.18 6.88 10.85 10.23 Extract. — The extract, or residue, represents the sum of the non- volatile constituents, including sugar, nitrogenous matters, tartaric and other acids, mineral and organic salts, coloring and astringent prin- ciples, glvcerin, etc., all of which are jiresent in but small quanti- ties. In swc^et wines, the princij)al constituent of the residue is sugar. The actual food value of the residue is, apart from the sugar, practi- cally vil. Adulteration of Wines. — ^^'ines have been subject to a wide variety of adulterations from the earliest times, and measures against the practice of their sophistication Avere enibrced long before those against the adulteration of bread and other foods Avere thought of. The ancient Greeks and Romans, for example, enacted stringent laws and a]ijiointed officials whose duty was to detect and })unish tho.-^e who offended. At the present time, adulteration of wines is practised very exven- sively, and includes the addition of water, of coloring agents, of pre- servatives, of glycerin to impart sweetness and body, of alum to heighten color and of decolorizing agents to remove it, the substitution of wholly artificial compounds, and processes for the " improvement " of the natural product. The flavoring and coloring agents are as a rule quite harmless. They are employed chiefly in the manufacture of factitious wines, and not uncommonly the same agent serves in both capacities. Prunes, raisins, dried apples and peaches, and dates are commonly so employed. Various berries, logwood, alkanet, red beets, coal-tar products, and a wide variety of other substances are said to be used for imparting color. WINES. 193 The addition of alcohol is recognized as a legitimate practice in the case of the fortified wines ; that of glycerin has no sanitary significance. The amount of alum used for heightening color is so small as to be productive of no harm. The employment of decolorizing agents is, like the substitution of artificial products, a fraud pure and simple ; but the use of preservatives, such as salicylic acid, formaldehyde, and sulphites, is objectiouable on account of danger to health. For the improvement of wines, a number of processes are in vogue. Chief of these is " plasteruig," which consists in the addition of gyp- sum to the must for the purpose of securing a more brilliant appear- ance and increasing the keeping qualities. This agent decomposes the potassium bitartrate, formiug tartrate of calcium and acid sulphate of potassium, which latter eventually is converted into the neutral sulphate. Ohaptalizing consists in the neutralization of the acidity of the must by the use of marble dust, and the addition of cane sugar or glucose. This process diminishes the natural acidity and increases the yield of alcohol. •Gallizing consists in diluting the must so as to reduce its acidity to a given standard, and adding a sufficient amount of cane sugar or glucose to insure the production of the proper alcoholic strength. The Pasteur treatment of wines is resorted to sometimes as soon as evidence of untoward fermentations producing the so-called " wine dis- eases" appears. The wine, best in the bottle, is heated to from 55° to 65° C according as the alcoholic strength is high or low, whereby the existing germs are killed and the preservation of the wine is made permanent. The manufacture of artificial wines is carried on extensively in this country and abroad, in spite of the fact that fair grades of the genuine product are obtainable at very low prices. A number of hand-books and guides to the "art of blending and compounding" are pub- lished for the use of wholesalers and retailers of wines and liquors, and from several of these the following are selected as examples of the methods given: (1) Port: cider, 30 gallons; alcohol, 5 gallons; syrup, 4 gallons ; kino, ^ pound ; tartaric acid, i pound ; port wine flavor, 6 ounces. (2) Claret : California hock, 40 gallons ; extract of kino, 8 ounces ; essence of malvey flower, 8 ounces. (3) Sherry : equal parts of Spanish sherry and California hock. (4) White wine : dissolve 25 pounds of grape sugar and 1 of tartaric acid in 25 quarts of hot water, add 75 Cjuarts of cold water and 50 pounds of grape pulp, stir, cover, let ferment for four or five days, and strain. In France, an artificial substitute for wine, known as " piquette," is manufactured very extensively from raisins and dried apples. It is estimated that in 1898 no less than 50,000,000 gallons were made and consumed. The process is exceedingly simple. To each gallon of water used are added 1 pound of raisins and 1 of dried apples ; the mixture is placed in an open vessel and allowed to stand three days. It is then bottled with ^ teaspoonful of sugar and a small piece of cinnamon in each bottle. It is said to be a pleasant and harmless beverage. 18 194 FOODS. Analysis of Wines. Determination of Alcohol. — The process for the determination of alcohol is the same as that followed in the analysis of beer, except that the distillation or evaporation is carried farther. At least 60, or better 75 cc, are collected by distillation or driven off by open evapo- ration. Determination of Extract. — The specific gravity of the de-alcohol- ized wine gives, as with beer, an approximate estimate of the amount of extract, and the same table may be used. The direct determination is made by evaporating 50 cc. of the wine in a weighed platinum dish on a water-bath and drying to constant weight in an air-bath. AVith sweet wines, a smaller amount is preferable. Determination of Acidity. — The total acidity, due to bitartrate of potassium, tartaric, malic, and other acids, is reckoned as tartaric acid. Twenty-five cc. of the wine are titrated in the usual way with decinor- mal sodium hydrate, 1 cc. of which equals 0.0075 gram of tartaric acid. The volatile acids are reckoned as acetic acid. Fifty cc. of the wine are placed in a distilling flask connected by means of its outlet tube with a Liebig condenser, and, by means of a bent tube passing through its stopper and projecting well below the surface of the wine, with a flask containing 250 cc. of water. The contents of both flasks are brought to the boiling-])oint, and then the flame beneath the wine is turned down and the current of steam passed through until 200 cc. of distillate are collected. This is titrated with decinormal sodium hydrate, and the result is expressed as acetic acid. The determination of the amounts of the individual acids is of no hygienic interest. Determination of Sugar. — The amount of sugar in wine is deter- mined l)y reduction of Feliling'.s solution, by the method of Allihn, and by polariscopy. For the details of these methods, the reader is referred to any of the standard works on wine analysis, for the small amount of sugar ordinarily present is of but little hygienic interest, and the description of the processes would require an amount of space vastly out of proportion to the importance of the subject. Determination of Ash. — The residue obtained in the determination of extract can be utilized for the determination of the ash. It should be ignited at as low a temperature as possible. Detection of Coal-tar Colors. — While the presence of coal-tar colors is not difficult of detection, the identification of the individual members of the group is by no means easy. The following tests give reliable indications of the presence of this class of colors. Equal volumes of wine and ether, agitated in a flask, and let stand and sepa- rate, will show in the ether layer a red coloration, if anilin colors are present. In place of ether, nitro-benzene may be used ; this removes fuchsiu, eosin, and methylen-blue, but does not take up any of the vegetable colors, safranin, or indigo-carmine. Amyl alcohol, also, will become reddened when agitated with wine containing anilins, but the WINES. 195 wine must first be made slightly alkaline. If white woollen threads are immersed for some time in the colored liquids, they will take up the colors and become dyed. Cazeneuve's test is performed as follows: To 10 cc. of wine add 0.20 gram of mercuric oxide, then shake for one minute, boil, let stand, and filter. The filtrate should be clear, and in the absence of anilins should be colorless ; if it is red, an anilin color is present. Absence of color is, however, not conclusive evidence of purity, since a number of the anilin colors, as eosin, methylen-blue, and others, are wholly precipitated, and so do not appear in the filtrate Safranin, methyl-eosin, Ponceau red, and a number of other colors are precipi- tated partially or completely. A number of these, including safranin, Bordeaux red, and Ponceau red, may be separated by the following process : To 200 cc. of wine from which the alcohol has been expelled, add 4 cc. of 10 per cent, hydrochloric acid and some wliite woollen threads, and boil for five minutes. AVithdraw the threads and wash them with cold water acidu- lated with hydrochloric acid, next with hot water similarly acidulated, and lastly with distilled water alone. Boil the threads in 50 cc. of distilled water containing 2 cc. of strong ammonia water, remove them, and immerse new ones. ]\Iake acid with hydrochloric acid and boil for five minutes. Varying shades of rose-red will be imparted to the threads if any of these colors are present. Fuchsin may be detected by the following methods : (1) To 100 cc. of wine add 5 cc. of ammonia water and 30 cc. of ether, and shake. Remove the ether, which will have no color, place it in a watch-glass with a white woollen thread, and let it evaporate to dnmess. If even a trace of fuchsin is present, the thread will show a distinct rose-col- oration. (2) Mix 2 volumes of wine and 1 of solution of basic ace- tate of lead, warm gently, and shake. Filter, add to the filtrate a small amount of amyl alcohol, shake again, and remove the amyl alco- hol. If this has a red color, it may be due to fuchsin or to orseille. To a portion of the colored liquid add hydrochloric acid ; if the color is discharged, it was due to fuchsin. To another portion add ammonia water ; if the color is changed to purple violet, it was due to orseille, Detection of Preservatives. — Salicylic Acid. — Spica's method for detecting saHcylic acid in wine is as follows : Acidify 10 cc. of wine with a few drops of hydrochloric acid, and shake with an equal volume of ether. Remove the ether, filter it if necessar}^, and evaporate to di'vness. Add a drop of nitric acid, warm gently, and add an excess of ammonia and 1 cc. of water. Immerse a white woollen thread, apply gentle heat, and then withdraw the thread, wash it, and dry it between pieces of blotting-paper. A yeUow color indicates that sali- cylic acid was present in the wine. Another method, for which great delicacy is claimed, even to a tenth of a milligram in a liter, is the following : Acidify 50 cc. of wine, beer, or other li(|uid with sulphuric acid, and shake it with an equal volume of a mixture of ecpial parts of ether and naphtha. Separate 196 FOODS. the ether, filter, and evaporate down to 5 cc. ; then add 3 cc. of water and a few drops of very dihite sohition of ferric chloride, and filter through a wet filter. In the presence of salicylic acid, the watery por- tion will have a violet color. A modification of this method consists in extracting with ether alone, and then extracting the ether residue with naphtha ; the residue on evaporation of the naphtha is treated with water and voiy dilute ferric chloride. Formaldehyde. — To 10 cc. of wine, add a few drops of milk known to be free from formaldehyde, and shake in a test-tube. Next pour down the side of the tube about 4-5 cc. of strong commercial sul- phuric acid, and note the color at the line of contact of the two liquids. (See under Milk.) Sulphites. — To 200 cc. of wine (or beer) add 5 cc. of phosphoric acid ; distil 100 cc, using a Liebig condenser with a bent delivery tube which dips below the surface of '20 cc. of decinormal solution of iodine. By distilling in a current of washed CO._,, the danger of back suction is avoided. The reaction which is brought about is as follows : SO, + 2H,0 + I, = H,SO, + 2HI. The amount of SO, may be determined by estimating the excess of iodine by means of standard sodium thiosulphate, or the distillate may be acidified with hydrochloric acid and the contained sulphuric acid precipitated as barium sulphate by the addition of barium chloride. One milligram of barium sul- phate is equivalent to 0.2748 milligram of SO,. CIDER. Cider, or apple wine, is the fermented juice of the apple. It is made very extensively wherever apples are grown, and is a very important product, viewed either as a beverage or as the basis of what is regarded generally as the best kind of vinegar. A very large, if not the greater, part of the cider produced is made without special care by a very simple process. The apples used are ordinarily those not marketable on account of small size, greenness, over-ripeness, or bruises ; but often perfect fruit is used when the crop is so abundant that there is more profit in converting it into cider and vinegar than in sending it in barrels to market. The fruit is ground to a pulp and pressed, and the juice is draAvn into barrels and allowed to ferment. If the same amount of care is taken as is given to the making of wine from grapes, the ])roduct is of a superior grade, and keeps very well ; but as ordinarily made in the country, its life is short, unless treiited with salicylic acid or other preservative to check fermentation. In France, where the yearly yield is very great, the best grades are made with due regard to the temperature at which the fermentation proceeds, and to the importance of racking off and fining. Cider of good quality contains usually from 8 to 5 per cent, and sometimes as much as 8 per cent, by weight of alcohol. Very new sweet cider may contain less than 1 per cent. The total extract, which is largely sugar, is in inverse proportion to the amount of alcohol ; in DISTILLED ALCOHOLIC LEVEE AGES. 19 i average samples, it amounts to from 4 to 6 ])er cent,, while in new sweet cider it is commonly nearer 9 per cent. The free acids, chiefly malic, amount to less than 0.75 per cent., and average about 0.40. The adulterants of eider are water and salicylic acid. The latter is found very commonly in that \\hich reaches the city markets. PERRY. Perry, or "pear cider," is the fermented juice of pears. It is made in the same way as cider. Pear juice Ijeing richer in sugar than apple juice, it follows that the average content of alcohol is somewhat higher than in cider. Distilled Alcoholic Beverages. Spirits, or distilled liquors, are the product of distillation of fer- mented sugar solutions. Their most important constituent is ethylic alcohol, which is ordinarily present to the extent of about 45 per cent. AVhen freshly made, they contain variable small quantities of higher alcohols, furfurol, fatty acids, and other volatile principles, which together constitute what is known as fusel oil, the chief constittient of which is amylic alcohol. Each kind of grain or other raw material from wliich the ferment- able sugar solution is obtained yields a diiferent kind of fusel oil ; dif- ferent because of the changing relative proportions of its constituents, which include btitylic, propylic, and amylic alcohols, and their corre- sponding acids, -Ijutyric, propionic, and valerianic, and other matters. That which is foimd in potato spirits is richest in amylic alcohol, and is the most toxic, while that from grapes contains by far the least and produces the least harm. During the process of aging, the constituents of the fusel oil undergo chemical changes which result in the formation of cenanthic, acetic, and butyric ethers, acetate and valerianate of amyl, and other compounds, which together constitute the aroma or " bou- quet." Thus, a spirit is improved in two ways by long storage : it loses in toxicity and gains in flavor. The relative toxicity of the several alcohols and of other constituents of fusel oil has been determined by Dujardin-Beaumetz and others, who show that the poisonous properties increase with the boiling-point and molecular weight. JefProy and Ser\'eaux ^ determined the amounts in grams necessary per kilogram to kill a rabbit, as follows : ethylic alcohol, 11.70 ; propylic alcohol, 3.40 ; butylic alcohol, 1.45 ; amylic alcohol, 0.63; furfurol, 0.24. Daremberg- found by experiment that artificial spirits and wines made ^^dth pure rectified alcohol are less toxic than the genuine products, by reason of the absence of the constituents of fusel oil. Poubinowitch," speaking of the greater toxicity of the higher alcohols, calls attention to the fact that the distillates from cider^ perry, and fermented grains, potatoes, and molasses, are much more toxic than brandy. ^ Archives de Medecine experimentale et d'Anatomie pathologique, 1895, p. 569. ^ Ibidem, p. 719. ^ Gazette des Hopitaux, 1895, p. 237. 198 FOODS. Most spirits are colored artificially by the addition of harmless col- oring agents, the most widely used of which is caramel. As the prac- tice of coloring is in response to the demand of the consumer for a darker color than can be obtained otherwise, it can hardly be regarded as an adulteration. BRANDY. Brandy is obtained by distilling wines of the poorer qualities, often mixed with the '^ lees," or dregs from the wine casks, and the " marc," or solid refuse left after pressing the grapes. The lees and marc are used also alone for the production of a higiily odorous brandy, which is much used for improving the flavor of other brandies, and for giving flavor to the artificial brandies made from pure alcohol and water. From tliis marc brandy is obtained the oily snbstance, oenanthic ether, which is known commercially as " oil of wine." Brandy is produced very largely in France, and much less exten- sively in Spain, Portugal, and Germany ; in California and in the wine- growing region of the Ohio and ]Mississi]ipi Valley, it is produced in large quantities and of most excellent quality. The colorless distillate is stored for some time in oaken casks, from which a small trace of tannin and a varying depth of amber color are acquired. The flavor, which in general de})ends upon the kind of grapes, their condition when pressed, and the care observed in the making of the wine, bect)mes improved during storage. The liquor is then colored and bottled for the market. , Good brandy should contain from 39 to 47 per cent, of alcohol by weight, should have an agreeable odor and taste, and should be free from substances added to imjwrt sharp taste and a])]xirent strength. The nearly dry residue from 100 cc. veiy slowly evaporated on a water- bath should have a pleasant odor, and its taste should be neither sweet nor sharp ; a sharp odor points to the presence of fusel oil derived from potato or cereals ; a sweet taste is indicative of added sugar or glycerin ; and a sharj) taste is suggestive of cayenne or other s])ice. Much of the brandy of commerce is a pm'ely artificial product made from alcohol or potato spirits, water, and flavorings. The formulse for making brandy are very numerous, and not a few require what is known as brandy essence, an article made with ethers and other substances in varying proportions. By one formula, it is made Avith 5 parts of oenan- thic ether, 4 of acetic ether, 3 of tincture of galls, 1 of tincture of pi- menta, and 100 of alcohol ; by another, it consists of 15 parts of acetic ether, 12 of sweet spirit of nitre, and 1 of rectified wood s]iirit. One part of either of these mixtures is sufticient to flavor a mixture of 1000 parts of alcohol and (JOO of water. As examples of the way in which factitious brandy is made, the fol- lowing Avill serve: (1) Boil ~) ounces of raisins and 6 of St. John's bread in water, filter, and make up to 10 quarts; mix this with 20 quarts of alcohol, 10 ounces of brandy essence, and h ounce of essence of violet flowers. (2) Dissolve 1 pound of argols and 3 of sug-ar in a WHISKEY. 199 gallon of water, add 40 gallons of alcohol, i pound of acetic ether, 2 ounces of tincture of kino, 6 pounds of bruised raisins, and a sufficient amount of caramel, and let stand for fourteen days ; strain and bottle. WHISKEY. AVhiskey is the product of distillation of the fermented mash of grain or potatoes. The raw materials from which the mash is made include malt, wheat, rye, corn, oats, and potato. In the process of mashing, the starch of the grain is changed to sugar by the diastase of the malt ; and since this ferment is capable of converting other starch thau that with which it is associated, it is customary to mix malt and raw grain in the proportion of 1 to from 5 to 9 parts. A bushel of grain makes about 2.5 gallons of spirits. In this country, the grains employed are chiefly corn, wheat, and rye ; in Great Britain, barley, oats, and rye are used together ; potatoes are used to a greater or less extent on the continent. The mash for Scotch whiskey is very com- monly prepared from 2 parts of malt, 7 of barley, and 1 each of oats and rye ; that for Irish whiskey is the same, with the exception of the rye. As soon as the fermentation of the mash through the affencv of veast is complete, the distillation is begun. The first distillate, known as ^' low wine," is re-distilled. The second distillate is stronger and less rich in fusel oil, which, being less volatile than ethylic alcohol, comes over chiefly in the later portions. The new whiskey is stored for sev- eral years, in order that it may acquire the flavor due to the formation of new compounds from the constituents of the fusel oil. During stor- age, it takes up a trace of tannin from the oak of the casks. The flavor of whiskey depends upon the nature of the raw material, and largely upon the aging process. The disagreeable flavor and odor of new whiskey are due to fusel oil ; the smoky taste of Scotch and Irish whiskies is due to the smoke of the peat and turf fires over which the malt is dried. Indian corn whisky has a much difierent flavor from that of rye whiskey ; this flavor is regarded highly by many to whom rye whiskey is unpalatable and insipid, and at the same time it is so full that to others it is rank and nauseating. The peculiar flavor of Bourbon whiskey, so-called because originally produced in Bourbon County, Kentucky, is due to the corn from Avhich, with rye, the mash is prepared. Whiskey of good quality should contain about 45 per cent, of alco- hol by weight, and should yield not more than 0.25 per cent, of resi- due, which should have a slightly aromatic odor and but little taste. Whiskey is manufactured very largely from alcohol, water, and various flavoring compounds, some of which can hardly be looked upon as wholly innocuous. The following directions are taken from a small work, the object of which is, according to the preface, " to give the dis- penser of liquors thorough and practical information by which he will be enabled to compound, and blend liquors for his own purposes, and thus secure the additional profit." 200 FOODS. 1. Bourbon Oil. — Take of fusel oil, 64 ounces ; acetate of potassium and sulphuric acid, each, 4 ounces ; and black oxide of manganese, 1 ounce. Dissolve ^ ounce each of sulphate of copper and oxalate of ammonium in 4 ounces of water, mix all in a glass percolator, and let rest for twelve hours. Then percolate and put into a glass still, and distil 64 ounces. 2. Rye Oil. — Mix 64 ounces of fusel oil, 8 each of oenanthic ether, chloroform, and sulphuric acid, and 2 of chlorate of potassium in 8 of water, place in a glass still, and distil 64 ounces. 3. Beading Oil. — Mix together 48 ounces of oil of sweet almonds and 12 of sulphuric acid, and when cool neutralize with ammonia and dilute with double the volume of proof spirit. " This is used to put an artificial bead on inferior liquors." For making the lowest grade of whiskey, one is directed to mix 32 gallons of alcohol and 16 of water, 4 ounces of caramel and 1 of beading oil. ^y adding oil of rye or oil of Bourbon, " making the result rye whiskey or Bourbon, as the case may be," the value is said to be increased. From another similar source the following recipes for factitious whiskey are taken : 1. Bourbon Whiskey — Proof spirit,^ 100 gjillons ; pear oil, 4 ounces ; pelargonic ether, 2 ounces; oil of wintergreen, 13 drachms in ether; wine vinegar, 1 gallon ; caramel color, a sufficient quantity. 2. Old Bourbo)!. — Alcohol, 40 gallons ; Bourl)on whiskey, 5 gal- lons ; sweet spirit of nitre, 2 ounces ; fusel oil, 2 ounces. Mix and let stand four days. 3. Old Rye. — Soak a half peck of roasted dried peaches, put them into a woollen bag and leach with common whiskey sufficient for a barrel, and add 12 drops of strong ammonia. 4. Scotch Wliiskcy. — Alcohol, 46 gallons ; genuine Scotch, 8 gal- lons ; water, 1 8 gallons ; ale, 1 gallon ; creasote, 5 drops in 2 ounces of acetic acid ; pelargonic ether, 1 ounce ; honey, 3 pounds. 5. Iri.sh Whi.^kcy. — Same as above, substituting Irish for Scotch, and omitting the honey. RUM. Rum is made by distilling fermented molasses or the skimmings of sugar boilers, with, not uncommonly, other substances, as pineaj)ple.s and guavas, to give flavor. The characteristic flavor of rum is due to butyric ether. The alcoholic content of rum is very variable, ranging from 30 to over 60 per cent, by weight. Like other spirits, rum is very largely an artificial product of alcohol, water, and flavorings ' Proof spirit is defined by an act of Parliament as a diluted spirit which at 51° F. shall weigh exactly twelve-thii'teenths as much as an equal measure of distilled water. It contains half its volume of alcohol of sp. gr. 0.7939 at fiO° F., or 49.5 per cent, by weight, or 57.27 per cent, by volume of absolute alcohol. Its sp. gi". is 0.91984. Over and under proof mean that a spirit is stronger or weaker than proof spirit, and the excess or deficiency is expressed as so many degrees over or under proof. The expres- sion, for example, 25 under proof, means that the specimen consists of 25 parts of water and 75 of proof spirit ; 25 over proof means that 100 parts may be diluted with 25 of water to bring it to the strength of proof spirit. LIQUEURS. 201 known as rum essence. One of these consists of 15 parts of butyric ether, 2 each of acetic ether, essence of vanilla, and essence of violet, and 90 of alcohol. Another consists of 32 parts each of rum ether and acetic ether, 8 of butyric ether, 16 of extract of saffron, and ^ of oil of birch cut in strong alcohol. The rum ether required is a product of the distillation of alcohol, sulphuric acid, pyroligneous acid, and black oxide of manganese. Prune juice is also a common addition to factitious rum for its flavor and color. GIN. Gin is an alcoholic liquor flavored with juniper berries and a great variety of other substances, including cardamom, coriander, cassia buds^ calamus, orris, angelica root, orange peel, licorice powder, and sugar. It should contain about 40 per cent, of alcohol, and not over 6 per cent, of total residue, including sugar. Liqueurs. Liqueurs, or cordials, are manufactured compounds of alcohol, essen- tial oils, cane sugar, and coloring matter. They contain usually about 40 per cent, of alcohol by weight, and from 25 to 50 per cent, of cane sugar. The colorings are, as a rule, of vegetable origin, but sometimes the coal-tar colors are employed. In the small amounts in which they are consumed at any one time, they can hardly be looked upon as espe- cially harmful apart from their alcoholic content. There is one, how- ever, very largely consumed diluted with water as a "long drink," that appears to exert a decidedly deleterious effect upon the nervous system ; namely, absinthe. The evil eifects of this drink are by some attributed to the oil of wormwood (^Artemisia absinthium), and by others to the star anise {lllieiwrn), both of which are among the numer- ous ingredients used in its manufacture. Where the blame lies is of no great consequence, the drink being one M^hich should be shunned above all others as a poison without regard to the innocuousness of most of its constituents ; but it is unlikely that its disastrous effects are due to wormwood, which as a drug has little or no action, and which enters into the composition of another drink, vermuth, which enjoys a good reputation. This is in no sense a cordial, but for convenience may here be described. Vermuth is a fortified white wine in which certain herbs and other vegetable matters have been infused. The ordinary French vermuth is made from wormwood, bitter-orange peel, water germander, orris root, chamomile, Peruvian bark, aloes, cinnamon, nutmeg, centaury, and raspberry, but many other substances are used by different makers. The fresh product has a very pronounced flavor, which is mellowed by age. The wines most used in making French vermuth are from the Rhone "Valley, Picpoul, and from the southernmost parts of France. Italian vermuth differs materially from the French ; it is a much 202 FOODS. weaker infusion with a far more bitter taste. The materials used are in the main the same, but they are employed in very different propor- tions. Vermuth contains about 17 per cent, of alcohol. Section 6. CONDIMENTS, SPICES, AND BAKERS' CHEMICALS. The condiments include a large number of food accessories which, while they are themselves of no nutritive value in the amounts which it is possible to eat, serve a very useful purpose in imparting flavor, and in stimulating appetite and digestion. Among them are some which act through free acids, some through volatile oils, some through resinous matters, and one, perhaps the most important of all, common salt, through itself alone. Some are simple substances ; as vinegar, salt, and the spices ; while others are combinations of a number of ingredi- ents blended according to definite and, as a rule, secret formulas ; as sauces, chutneys, catsups, and curries. Only when these compounded articles contain substances injurious to health can they be regarded as adulterated. The tomato catsups are preserved very commonly with salicylic acid or other preservatives, and colored with anilin dyes. Thus, of 25 samples of different makes examined in 1897 by the health authorities of San Francisco, 20 contained salicylic acid, 2 contained this agent together with borax, and 1 contained formaldehyde ; 16 were artificially colored, mostly with coal-tar colors. Of 39 examined by the ^Massachusetts State Board of Health during 1899, 15 contained salicylic acid and 13 benzoic acid. VINEGAR. Vinegar is a weak solution of acetic acid resulting from the acetous fermentation of saccharine solutions which have undergone alcoholic fermentation. It contains, in addition to acetic acid, small and unim- portant amounts of alcohol and aldehyde, and extractive matters in varying amounts, according to the nature of the original liquid. The acetic acid contained is the product of oxidation of alcohol through the agency of Mycodermn acefi, a fungus which forms what is known as the " mother of vinegar." Thus, the change from sugar to acetic acid involves two separate fermentative changes through the agency of two different organisms, Saecharomyces cerev'mcc and Mycoderma aceti. There are several kinds of vinegar in common use, as follows : Cider Vinegar. — In this country, cider vinegar is regarded very generally as the most desirable kind. It contains no aldehyde, about 4.50 to 5.50 per cent, of acetic acid, marked traces of malic acid, and about 2 per cent, of total residue, or "cider-vinegar solids." Wine Vinegar. — In wine-]>roducing countries, the vinegar in com- mon use is made from the cheaper kinds of wine. It has color or not, according to the kind of wino from which it is made. The so-^^alled white wine vinesrar in common use in this countrv among- the foreign- VINEGAR. 203 born population is a colorless product of the oxidation of dilute spirits. Wine vinegar contains rather more acetic acid than cider vinegar, but far less residue. Malt Vinegar. — In England, which is neither a cider-j)roducing nor a wine-producing country, the vinegar in commonest use is made from a wort prepared from malt and unmalted grain. It is less strong in acetic acid than the vinegars already described, but commonly con- tains sulphuric acid, which, under the English law, is a permissible admixture to the extent of not exceeding 0.10 per cent. Molasses Vinegar. — A very large part of the domestic supply of vinegar is manufactured from fermented molasses. It is made to imitate cider vinegar in color, and is sold commonly under the name •of that article. It yields about the same amount of acid, but is very •deficient in residue. The latter has a very bitter taste, and after com- plete ignition yields an ash containing no potassium salts, while that from cider vinegar gives a decided indication. Spirit Vinegar. — Spirit vinegar, also known as " Quick Process " vinegar, is made from diluted alcohol. The process used is the same as that employed in the making of malt vinegar and molasses vinegar. A series of suitable vats is constructed and filled with beech or birch shavings or twigs, which by appropriate treatment become coated with Mycoderma aeeti. The alcoholic liquid is allowed to percolate through, and in its passage the alcohol is transformed. The temperature of the room is maintained at about 70° F. Adulterations of Vinegar. — The principal adulterations of vinegar are the addition of water and the coloring of inferior grades so that they may be sold as cider vinegar. Where laws are in force establish- ing standards of acidity and residue for cider vinegar, a very common fraud is the addition of cider jelly or of a preparation made from apple pomace to a cheap vnnegar of the proper strength, colored, if necessary, with caramel. Such compounds always show but slight or no reaction when tested for malic acid. The addition of mineral acids is not a •common practice in this country. Examination of Vinegar. — Acidity. — To 6 cc. of the specimen in a porcelain casserole, add a few drops of phenolphthalein solution and about 20 cc. of distilled water. Titrate with decinormal sodium hydrate solution, adding little by little until the appearance of a faint pink coloration. The number of cc. of the reagent used, divided by 10, equals the percentage of absolute acetic acid. Residue. — Evaporate 5 grams in an accurately weighed platinum dish to complete dryness over boiling water. After the residue is weighed, it may be ignited for its yield of ash. Genuine cider vinegar should give no more than a faint cloudiness on being tested with nitrate of silver and chloride of bariimi (absence of more than traces of chlorides and sulphates), and should yield a fairly copious precipitate with solution of subacetate of lead (presence of malic acid). The residue should not taste bitter (absence of caramel). Cider vinegar to which water has been added is likely, according to 204 FOODS. the nature of the water, to show more thau the usual results ou test- ino; for chlorides and sulphates, and to yield notable traces of lime. Molasses vinegar generally yields marked indications of lime salts and a more or less pronounced odor of rum. LEMON JUICE AND LIME JUICE. Lemon juice is the expressed juice of the ripe fruit of Citrus limonum. It is a somewhat turbid yellowish liquid, with a very acid taste and a slight agreeable odor, due in part to the presence of a small trace of volatile oil from the rind. It should contain about 7 to 10 per cent, of citric acid, aud should yield from 0.50 to 1.00 per cent, of ash. Its specific gravity should be not less than 1.030, and is usually above 1.040. As it is quick to undergo decomposition in its natural con- dition, a number of methods have been proposed for its presers-ation, the best of which appears to be, first to clarify it by means of strong alcohol, next to filter or decant from the precipitated matters, and then to expel the alcohol by heat. The clear juice may then be bottled and sterilized. Lime juice is the expressed juice of the sour lime, Citrus acida, and of the sweet lime, C. Urnetta. It contains usually somewhat less acid thau lemon juice, and has a lower specific gravity. It is preserved by the same method. As antiscorbutics, lemon juice and lime juice are of about equal value, aud far superior to vinegar. Adulteration. — Lemon juice is nuich more subject to adulteration thau lime juice, but both are ialsified aud imitatetl extensively. In fact, it would not be overstating the case to say that by far the larger part of the lemon juice sold in this country is wholly factitious. Com- monlv, it is nothing more than an aqueous solution of citric acid ; some- times, it is flavored with oil of lemon. Its taste is much shar])er and less agreeable than that of tlie genuine article. The residue is very different in character and appearance, and leaves practically no ash on ignition. Other acids are used sometimes in place of or in addition to citric acid. The one most commonly emj)loycd is said to be tartaric ; this is detected readily by the grachial formation of bitartrate of potas- sium on addition of tJie acetate. The mineral acids are said to Ix' added not infrequently ; they are detected without difficulty by the common tests. SALT. The best grades of common salt arc white, dry, free from dirt, and completely soluble in water. iSIany s|)ecimens of good quality contain traces of chloride of magnesium, which causes caking. In humid weather, even the best grade of salt absorbs moisture sufficient in amount to cause it to lose its dry, powdery nature. The addition of about 10 per cent, of corn starch serves to keep it dry and pow- dered. CLOVES. 205 MUSTARD. Mustard is the flour of the seed of the black and the white mustard, 4Sinapis niger and S. alba. The first mentioned is much the more pun- gent of the two ; on being wet with water, a volatile oil is developed from two of its constituents. The white mustard yields no volatile oil by this treatment, but develops an acrid principle. Both varieties of seeds contain a bland fixed oil to the extent of 20-25 per cent. As this adds nothing to the flavor, makes grinding more difficult, and exerts an injurious influence on the keeping qualities, it is removed from the whole seeds by pressure. Mustard is largely subject to adulteration with wheat, rice, and corn flour, with the farther addition of turmeric to restore the color lost by dilution. These substances are detected very easily by means of the microscope. Furthermore, since starch is wholly absent from pure mustard flour, if a small portion of a suspected sample, boiled in a little water in a test-tube and cooled, gives a blue or bluish-black color on the addition of compound iodine solution, it unquestionably is adulterated. PEPPER. Pepper is the fruit of Piper nigrum, a perennial climbing shrub. The imripe berries, dried for several days after being picked, are known as Black Pepper. The ripened berries, dried and decorticated, are known as White Pepper. In the powdered form, in which they are retailed most commonly, both are adulterated very extensively with substances of a harmless nature. These include ground shipbread, cornmeal, cocoauut shells, rice, buckwheat, oatmeal, mustard hulls, charcoal, olive stones, and a variety of other substances of little or no value, capable of being reduced to powder. The simplest method of determining the purity of this or any other ibrm of spice is to reduce a specimen of the genuine unground sub- stance to powdered form and study its appearance under the microscope, and then to compare it with the sample in question. Each kind has its characteristic appearance, and so with a little practice one is enabled to determine very quickly the question of purity. By a similar study of the microscopic appearances of the common adulterants, these may readily be identified in the mixture. The chemical analysis is intricate and tedious, and not always conclusive. CLOVES. Cloves are the flower buds of Eugenia caryophyllata, picked while red and dried in the sun. They contain about 16 per cent, of volatile oil, easily removed and of considerable value. In the powdered form, cloves are adulterated commonly with allspice, clove stems, spent cloves, cocoanut shells, and other worthless matter. The presence of ^pent cloves can be determined only by estimation of the amount of volatile oil present. Clove stems show microscopically a very large 206 FOODS. proportion of the so-called stone cells. Other substances are detected in the manner described vnider Pepper. CINNAMON AND CASSIA. Cinnamon is the inner bark of Cinnamomum zeylanicum. Cassia is the bark of several species of Cinnamomum. In the unground state, cinnamon is thin and delicate ; cassia is thick and comparatively coarse. Cinnamon is the richer in volatile oil, and for this reason and because it is found much less abundantly, is considerably more expen- sive than cassia. Ground cinnamon is practically never found in the market, the substance sold under that name being almost invariably cassia. The common adulterants of cassia include ground shipbread, nut shells, and cedar sawdust. ALLSPICE OR PIMENTO. Allspice is the dried imripe berries of Fimenta officiiiaVm. Although one of the cheapest of spices, it is adulterated extensively with ground shipbread, charcoal, nut shells, clove stems, and mustard hulls. GINGER. Ginger is the rliizome of Zinyibcr officinale. It is one of the most commonly adulterated of condiments. The substances used include ground shipbread, rice, mustard hulls, cayenne, turmeric, commeal, clove stems, and exhausted ginger from the manufacture of the tincture. It is very rich in starch, which is differentiated easily from other starches. NUTMEG. Nutmeg is the inner kernel of the fruit of Myridka fragrans. It is not commonly sold in tlie powdered condition, but when so sold is generally adulterated with the substances used as admixtures of other spices. MACE. Mace is the dried membranous covering, the arillode, of the nutmeg. It is adulterated with wild mace, cornmeal, and other cheap materials. CAYENNE PEPPER. Cayenne is not a true pepper, but the powdered pods of several species of Capsicum, including C. cmuuion and C. fasti giatum. Its appearance under the niicrosc()])e is very characteristic. The common- est adulterant is cornmeal. Among others are rice, mustard hulls, turmeric, and ground sliipbread. BAKING POWDERS. Baking powders, like condiments, arc in no sense foods, but being emj)loyed in the preparation ut up separately or combined in the form of soups, are jiractically un- damageable by any climatic heat " provided they are of the best quality and have been properly cooked and enclosed in perfectly sound air- tight tins. "Given these conditions, nothing can be more admirable; failing them, nothing more deleterious." In military oj)crations in the ti'o])ics, where beef cattle cannot be taken along on the hoof and re- frigerated beef cannot be transported overland on account of speedy decomposition, canned meats are indispensable. FOOD PEESERVATION. 211 How long: properly canued foods will remain in good condition, can hardlv be determined, but the evidence at hand points to indefinite preservation. In 1824, according to Letheby, a number of tins of mutton were cast ashore from the wreck of a ship at Prince's Inlet ; eight years later, they were found by Sir John Ross, and those which he opened were in good condition, although exposed during this time to alternate freezing and thawing. Sixteen years afterward, they again were found by men from the ship Investigator ; and in 1868, forty-four years from the time they were cast ashore, the remaining tins, opened by Letheby, were found to be in a perfectly sound state. Tyndall ^ makes mention of tins in the Eoyal Institution that had remained in good condition sixty-three years. Professor A. H. Chester, of Hamilton College, relates that in the summer of 1875 he hid a number of cans of corned beef under a stump in woods in the northern part of ^Minnesota, and five years later found them to be perfectly sweet, although they had been exposed to the heat and cold of five successive summers and winters. Again,, a number of cans of meat and fruit, washed into the Genessee Piver in 1865, were dug out of the mud sixteen years later, and found to be unaltered. It is an unfortunate fact that the cupidity of some of our largest packing-houses has led to the canning of what is practically refuse meat, from which the constituents to Avhich the desirable flavors are- due have been extracted, and that in consequence the public mind has largely become imbued with a prejudice against canned meats in gen- eral. But it is not alone in this country that canned meats are some- times not what they pur|)ort to be. It is related that in France, in 1899, a packer of meats was sentenced to pay a fine and to serve eight months in prison for putting upon the market an immense amount of canned game and poultiy, all of which had been made from the flesh of broken-down cab horses. Chemical Treatment. — Chemical preservatives are substances or combinations added to foods with the object of delaying or preventing their decomposition. They are used on the assumption that, while they accomplish the desired object, they are incapable of exerting any harmful influence upon the system of the consumer — an assmnption that has not been demonstrated as based on sound reasoning. It is assumed that bad eifects cannot be caused, because they are not mani- fested at once after the ingestion of small doses by persons in good health ; but this is no proof that continued use may not result in serious trouble which may be referred to some other possible cause. It is said that the preparations employed are in common use as valuable remedies in the treatment of the sick ; but it should be taken into consideration that, when used as remedies in morbid conditions, they are given for only a limited time, for the purpose of counteracting abnormal influences, and that the doses are regulated carefully under proper professional supervision. Their action in conditions of health 1 Floating Mattei-s in the Ah; New York, 1882, p. 293. 212 FOODS. and disease may be veiy different ; but whether so or not, one can find no excuse for the ingestion of curative remedies by a person in a state of health, whose system needs no such aid, for indefinite periods and with no regulation of the size of the dose. Salicylic acid, for example, is a remedy holding a high position in the treatment of rheu- matism, but its value in this condition is no valid excuse for its admin- istration day in and day out to those who never have felt the twinges and pain of this disease. It is much more reasonable to assume that drugs which exert a powerful influence for good in morbid states will exert an equal degree of influence for harm in conditions of health. Moreover, it is to be considered that the object of chemical treatment of foods is not to benefit the unconscious consumer, but to bring the largest possible financial return to the manufacturer and purveyor, to whom the health of the consumer may be a matter of little concern. In all fairness to the consumer, chemically preserved foods should be so labelled that the purchaser may be informed of the nature and amount of the added substance, so that tliose who object to the dietetic use of drugs may not have the same forced upon them without their knowledge. The addition of preservatives to foods offered for sale is forbidden in almost all civilized countries, and several governments have enacted laws specially directed against individual (hnigs. Thus, France names boric acid, borax, salicylic acid, and sodium bisulphite ; Austria names sali- cylic acid ; Germany prohibits all antiseptics, and especially boric acid and borates, and imposes additional ])enalties for the sale of chemically preserved foods to the navy. ^Massachusetts prohibits all preservatives except salt, sugar, niter, vinegar, and alcohol, unless the ]nn"chaser is informed of the nature of the substance used. In milk, all preserva- tives whatsoever are prohibited unconditionally. The substances used as chemical preservatives include boric acid and borax, salicylic acid, sul})hurous acid, sulpliites and sul])hatcs, benzoic acid, formaldehyde, hydrogen peroxide, sodium fluoride, and others of minor importance. Many of tlie commercial preparations in common use are combinations of two or more of these and other sub- stances. Thus, Venzke and Schorer * report the ingredients of 3ri('ht iiber T>ehonsmittel, etc., 1897, IV., p. 218. ^ Oestcrroiclu' Clu'inisclie ZiMtiin^', 1S98, I. * Giornalu dclla K. Socit'tri Italiaiui d'igiene, January 30, 1900, p. 1. CONTAMINATION OF FOODS BY METALS. 223 has been done properly, no trace of lead will be taken up by acid foods, such as tomato soup, or even vinegar ; but he advises that all new vessels should be cleansed very carefully before using, on account of the common presence of lead dust on the glaze when fresh from the kiln. Riche ' also determined that with properly fired ware the danger of solution of lead is practically nil, the specimens used yielding no traces to boiling dilute acetic and nitric acids and salt solutions. But im- properly fired ware will yield traces. Enamelled ware is believed commonly to contain lead ; and the enamel, having a diiferent coefficient of expansion from that of the iron, being likely to crack and chip off, especially with careless hand- ling, is thought to be dangerous ; but Barthe ^ found no trace of lead in a number of enamels examined, and asserts that very hard enamels need neither lead nor any other poisonous compounds in their prepara- tion. This accords with the experience of the author and other American investigators, and it may confidently be said that the enamelled ware in common use is lead-free. Aluminum ware, which has of late come into extensive use, is less acted upon by acid foods than tin, but is affected considerably by alkalies, the impurities present in commercial aluminum acting as favoring agents to its corrosion. But the resulting compounds are innocuovis in the small amount ingested. This kind of ware is kept clean very easily and offers the great advantage of lightness. Nickelware is attacked but slightly by ordinary food materials, and the amounts taken up are without sanitary significance. But its cost is against its extensive use, and, moreover, it imparts sometimes a greenish tint, which is repugnant to the eye. 1 Kevue d'Hygiene, August 20, 1900, p. 704. ''' Journal de Pharmacie et de Chemie, 1898, p. 105. CHAPTER 11. AIR. Air is a mixture of gases, and not a chemical compound. Until the latter part of the seventeenth century (1669), it was supposed to be an element, but Jean Mayow then proved it to be a mixture of gases ; and later, Lavoisier discovered the two gases, oxygen and nitrogen, which, a hundred years later, were separated by Priestley and by Scheele. Air is a colorless and apparently odorless mixture of oxygen, nitro- gen, argon, carbonic acid, aqueous vapor, and traces of other substances. It is not, however, under ordinary conditions odorless, but, on the con- trary, it contains various scents, to which we are so accustomed that, unless ]iresent in unusual degree, owing to local conditions, they are not perceived. This is noticed on returning to an ordinary atmosphere from one where the causes of the usual odors are absent or nearly so, as, for instance, from deep subterranean caves ; or from a room where the air is foul and oppressive, as, for instance, from a heated, over- crowded hall or street car. The air of the Arctic regions contains but little odor, on account of the absence of bodies which give rise to odors, and the proximity of any source of smell is noticed quickly. The ex- plorer Nansen ^ speaks of the pervading smell of soap which he noticed when, after months of wandering, he met Jackson, mIio had been housed comfortably with all the common necessities of man. While air is a mixture of gases, it is one of tolerably constant com- position, particularly in the case of its chief constituent, nitrogen. Under the conditions of life, the more important, but less abundant element, oxygen is subject to more or less variation. In the ])resence of vegetable life, particularly by day, it is increased slightly ; in the presence of animal life, it is diminished more or less. OXYGEN. The normal amount of oxygen is stated usually at just below 21 per cent, by volume. A. Leduc gives it at exactly 21, with 78.06 of nitro- gen and 0.94 of argon. DiflPereut observers have reported the follow- ing as averages of large numbers of analyses of pure outdoor air : 20.99 Scotland. 20.98 Scotl:ui(l. 20.94 Sweden. 20.92 France. " 20.94 Gernianv. 20.92 Nonvav.' 20.95 England. 20.95 Ohio. 1 Farthest North, Vol. II., p 529. 224 OXYGEN. 225 The mean of a number of analyses by Bunsen was 20.924 by yolume, and of a hundred at Paris by Eegnault, 20.960. For the sake of conyenience, we may disregard the yery slight difference between 21 and the figures obtained by exact analysis, a diiference in the second place of decimals, and accept 21 as a normal. At great heights, the proportion of oxygen is less than at the surface. For instance, on the Faulhorn, in Switzerland, 20.77 has been obseryed as the mean of a number of determinations. Under certain conditions, there is yery slightly more than 2 1 parts ; for instance, in the immediate yicinity of yegetation, especially by day, there may be an excess of oxygen, but it is yery small; sea air, taken in mid-ocean, has yielded 21.59, but ordinarily contains less than 21. It is less, by very small fractions, in the streets of cities than in the open country, and in towns than at sea. Oxygen is the element in air that supports all life. It is constantly being withdrawn from the air in the process of respiration, and is re- turned to it in chemical union with carbon as carbon dioxide. This is absorbed by vegetation and split up, the carbon being retained, and the oxygen for the most part released and returned to the air. Thus, the processes of animal and vegetable life combine to maintain the equilib- rium. All animals do not breathe in the same degree ; birds have the most active respiration, and next come mammals ; and all consume more oxygen when active than when asleep. Oxygen is essential to the germination of seeds, and to the growth of plants. Although plants take up carbon dioxide and exhale oxygen, thev also breathe as do animals, absorbing; the latter and exhalino- the former. Even the anaerobic organisms consume oxygen, although living where air is wanting, for they split up combinations of oxygen and other elements. Thus, in dilute sugar solutions they withdraw some of the oxygen and split up the sugar into carbon dioxide and alcohol. For sustaining animal life, it is essential that the air shall contain not far from the normal amount of oxygen ; that is, that it shall be neither much diminished nor yet over-rich in that element. Human life is impossible in air which contains but four-fifths of the normal amount, and equally so in an artificial atmosphere containing materially more than the normal. In man and animals, the tissues do not receive oxygen in the free condition, for when the air is inspired, the oxygen is taken up by the red blood corpuscles and unites with the haemoglobin to form an un- stable compound, oxyhjemoglobin, which, as the blood circulates through the tissues, is decomposed ; the oxygen is then taken up by the cells, and eventually returned to the blood in the form of carbon dioxide, and eliminated as such from the body. In an artificial atmosphere con- taining an excessive amount of oxygen, the haemoglobin becomes sat- urated with the gas, part of which becomes dissolved in the blood serum, and then acts as a poison to the tissues and destroys them. 15 226 AIR. Inspired air loses about a fourth of its oxygen, and is returned to the atmosphere rich in impurity ; but diffusion occurs so rapidly that the atmosphere of a thickly settled city shows no very material varia- tion from that of the open country. The lungs are never filled with pure air after the first respiration at birth, since they are never wholly emptied, and they consequently contain an impure residue of air after each expiration. The upper part of the respiratory tract is the only part that receives strictly pure air. Professor Richet has demonstrated that, if the respiratory tract be lengthened artificially by means of a rubber tube, pure air will never reach even the upper air-passages, and the animal Nvill die of asphyxia. The amount of oxygen absorbed varies with age, condition of health, and activity. According to Professor Foster, an average person inhales in 24 hours about 34 pounds of air, which corresponds to a little more than 7 pounds of oxygen ; and as the lungs absorb about a fourth of the oxygen inhaled, it appears that the average amount of oxygen absorbed daily is nearly 2 pounds. NITROGEN. The principal constituent of the air, nitrogen, takes no part in respiration, and is not mcreased in expired air ; but although it is indifferent and inert, it is, nevertheless, by no means unimportant. In the first place, it serves to dilute the oxygen, so that the latter is respirable ; and in the second place, it plays an imjiortant jiart in the growth of plants, the original source of all nitrogenous food, for that which we consume in the form of meat is from animals that have built up their tissues from vegetable food. As a diluent of oxygen, it serves to prevent too great activity of that element, which cannot be breathed with impunity for any length of time when present in an atmosphere to a greater extent than its normal amount. How nitrogen is absorbed by plants, we know only in ])art. Certain low forms (mycelia, etc.) seem to absorb it directly from the atmosphere when exposed freely to light and air. Some of the higher forms (peas, beans, clover, etc.) acquire it through the agency of certain micro- organisms which are present in nodules in their roots, and without which they will not thrive. These micro-organisms take the nitrogen from the atmosphere and give it in some form to the plants. That this is so, is proved by the fact that the plants will thrive in a soil quite free from nitrogen (in clean sand, for instance), and store uj) in their tissues an amount of nitrogen far in excess of that which was originally pres- ent in the seeds, provided these micro-orgjinisms are present in the nodules of the roots. If they are not present, the plants will not thrive, but may be made to do so by the application of water contain- ing cultures of the organisms. Of the doubtless many species which can fix atmospheric nitrogen, or Avhich aid in doing so, the following may be mentioned : B. megatherium, B. jinorescens Uquefaciens, B. pro- teus vulgaris, B. butyricun, B. mycoides, B. mesentericva vulgatus. CARBON DIOXIDE {CARBONIC ACID). 227 On the other hand, certain plants, grown in the open air in soils free from nitrogen, and protected from receiving ammonia and nitrates from the rain, will show no more nitrogen in their whole organization than was present in the seeds from which they sprang. The subject is one which has been investigated but partly, and future research will doubtless show that, under natural conditions, all plant life takes up in some way more or less nitrogen from the atmosphere, as well as from nitrogenous compounds in the soil. ARGON. Up to the time of its discovery, the element argon was included under nitrogen in the tables of composition of the air. How much is present, is not yet accurately determined. It was discovered in 1894, by Lord Rayleigh and Professor Ramsay, by whom later it was esti- mated as composing about 0.75 per cent, of the atmosphere. Leduc gives its amount as 0.94, and Schloesing as 0.84. It is quite inert, and cannot be made to combine with any other element, although it has been combined by Berthelot with benzene under the influence of electric discharge. HYDROGEN. According to the extensive researches of Armand Gautier, hydrogen is present in sea air and other pure air in fairly constant amount — about 0.015 per cent. It is believed to be due to various fermentative proc- esses, and to be contributed also by mineral springs and volcanoes. So far as known, its presence is devoid of sanitary importance. Other elements, as krypton, neon, and metargon, also discovered by Professor Pamsay, coronium, discovered by Nasini, and several others, are interesting solely from a purely scientific standpoint. CARBON DIOXIDE (CARBONIC ACID). All air contains carbon dioxide as a constant constituent. The nor- mal average amount in pure air is but slightly in excess of 3 parts in 10,000, or about 0.03 per cent., and not 4 parts, as commonly is stated. As little as 2.03, and even 1.72, has been observed in the air on moun- tain-tops, although generally we expect more rather than less than the normal amount at high elevations. Carbon dioxide is a result of oxidation of organic matter, and owes its presence in the atmosphere to respiration, fermentation, combustion, and chemical action in the soil. An average man exhales about 20 liters in an hour, and very nearly a kilogram in a day ; women exhale less, and children and aged persons still less. The amount exhaled is increased by muscular exertion and diminished by rest. Birds send out more in proportion to weight than other animals. The respiration of millions and millions of human beings and animals is constantly throw- 228 AIB. ing into the atmosphere countless tons of the gas ; every ton of coal in burning yields more than 67,000 cubic feet; every cubic foot of coal gas yields about double its volume ; every pound of candle nearly three times its weight (2.769); every gallon of oil and kerosene, and every piece of wood used as fuel, contributes its proportion. Huge volumes are sent forth continually by the soil air, which contains it in abundance, and by mineral springs, the waters of which contain it under pressure. It has been estimated that, from all sources, 5,000 million tons are discharged annually into the atmosphere. It is slightly more abundant in cities than in the country, and at night than by day. It is highest in amount at a given location during autumn, and lowest in winter. It is more abundant inland than on the coast. It increases somewhat as we ascend from sea-level — according to Schlagintweit, up to 11,000 feet. Its removal from the atmosphere is mostly through the agency of growing vegetation, but materially also by absorption by bodies of water, which, at ordinary temperature and pressure, will take up its own volume of the gas. It has been calculated that the ocean contains about ten times as much as the whole atmosphere. All green plants absorb it by day, and by means of their chlorophyll break it up into carbon and oxygen, the former being used in building tissue, and the latter returned to the air as a waste product. This process of nutri- tion goes on only under the influence of light, and consequently by day ; but there is also a respiratory function that is active both day and night, and has the same eifect on air as that of the respiration of animals ; namely, the consumption of oxygen and discharge of car- bonic acid. But the respiratory process has but a trivial influence in comparison with the chlorophyllian function. It is estimated that an acre of Avoodland withdraws in one season about four and a half tons, retains more than one and one-fifth tons of carbon, and returns three and a quarter tons of oxygen to the air. The slight increase at night is due supposedly in part to its exhalation by plants in their respira- tion, and also to currents of soil air, which ascends as soon as the air at the surftTce becomes colder, and consequently heavier, than itself. During the day, the soil air is colder and remains stationary. C'arbon dioxide is a heavy gas, incapable of sup})orting combustion or respiration, and serving no useful purpose in animal tissues. It constitutes about 4 per cent, of expired air, in which it is an excretion of the body. It is in itself inert, and incapable of exerting any poi- sonous action, but will cause asphyxia when present in sufficient amount to interfere with the atmospheric oxygen in the performance of its function. An atmosphere of respired air, containing 4 per cent., of carbon dioxide and about 16 per cent, of oxygen, will not support life longer than a short time, since the blood cannot get sufficient oxygen for the needs of the cells and tissues, and, in addition, cannot rid itself of its CO^. Gas exchange between the blood and inspired air depends upon the tension of the gas in both media, and, therefore, as soon as the tension of the CO2 in the atmosphere exceeds that of the COg of the blood, the OZONE. 229 blood corpuscles cannot excrete it, but must retain it. In consequence, asphyxia occurs. The question as to how much CO2 is permissible in air, has been an- swered variously. We assume 3 parts in 10,000 as the normal amount, and all in excess as impurity due to respiration and combustion. A total of 6 or 7 parts in 10,000 is regarded by the best authorities as the permissible limit, and 10 in 10,000 as distinctly harmful. AVhen the amount reaches 10 parts in 10,000, the air begins to be "close" ; and when it reaches 15 in 10,000, it is likely to cause headache m those unaccustomed to impure air. In crowded assembly rooms, as churches, theatres, and schools, the amount may reach 100 parts in 10,000 ; and more than twice as much has been found in a Swiss stable crowded with men and animals. The air of the hall in which the German Public Health Society (Deutscher Verein fur oflPentliche Gesundheits- pflege) met in Niiremberg in October, 1890, contained 24.10 in 10,000 at the beginning of one of the addresses, and 43.20 at its close. A large amount of carbon dioxide may be present in air without producing any ill effects, if there is plenty of oxygen present. Thus, Hegnault and Reiset have proved that animals can live in a mixture of 25 per cent, carbonic acid, 30—40 per cent, oxygen, and nitrogen. It has been held generally that CO2 up to 20 : 10,000 is in itself harmless ; that the deleterious agents in polluted air are organic mat- ters thrown off by the skin and lungs in company with it ; and that it serves as a convenient index of their amount. It has been the custom to say, " The more carbonic acid we find, the more organic matter we infer." These poisonous organic matters, however, though much sought after, have never been isolated, although a number of observers, using faulty methods, have from time to time obtained erroneous results. This subject will be considered farther on. OZONE. Ozone is a normal but by no means constant constituent of the air. It is generally absent from the air of large towns and cities, and is almost never present in the air of an inhabited room or near decom- posing matter. It is found in minute amounts (maximum, 1 : 700,000) in the open air of the country and sea. It is most abundant at sea and near woods, and somewhat more abundant on mountains than in valleys and on plains. It is more abundant in the morning in the colder months, and in the evening during hot weather ; it is more abundant in winter than in summer. It is stated that it is most abundant di- rectly after a thunder-storm, but beyond the fact that it is produced by the passage of the electric spark, there is nothing to substantiate this statement. As a matter of fact, the origin of ozone in the atmosphere is unknown. Ozone is an allotropic form of oxygen, consisting of molecules con- taining three atoms of that element. It has been liquefied under great pressure (127 atmospheres), and in that condition, and, indeed, in the 230 ATR. gaseous form, has a deep-blue color. It is produced by the passage of the electric spark, by slow oxidation of phosphorus, and in the electrol- ysis of wat^r ; but, as has been said, its origin as a normal constituent of the atmosphere has not been explained satisfactorily. It has an odor not unlike that of diluted clilorine. It has very strong oxidizing power, much more so than oxygen, which it exercises most actively both on metals and on organic matter ; hence its absence from the air of inhab- ited rooms and of densely populated areas, charged with organic matter and dust of all sorts, is easily explainable. To this property, its dimi- nution in autumn, when decomposition products are generated most actively, may properly be attributed. Its presence in the air of any place is fair evidence of freedom from oxidizable matters. Ozone has an exceedingly irritating effect on the respiratory mucous mcml)ranes, and when inhaled with oxygen in the proportion of 1 ])art in 240, quickly produces death in animals subjected to it. It is be- lieved to exert a pernicious influence in inflammatory conditions of the lungs and bronchi, even when present in not much more, if any, than the ordinary amount in the atmosphere. AVe actually know little or nothing of the effects of ozone on the svstem in the amounts ordinarilv present in air, but the absurdity of the expression so often used, that one has " gone to breathe the pure ozone " at a health resort is manifest. Peroxide of Hydrogen (H^Oo) is believed to exist in minute traces in the atmosphere, and to exert some influence in the process of oxidation. AMMONIA. Ammonia is constantly present in the air in very slight traces. It exists in the free state and in combination as nitrate and carbonate. Daily analysis of the air at ^Slontsouris for rive years gave as a mean for ammonia 2.2 milligrams j)er 100 cubic meters. It proved to be highest in amount in summer and lowest in winter. It is diminished in rainy weather, because it is absorbed by the rain during passage through the atmosphere ; it is increased Avith rising temperature some time after rain has ceased falling. As it is one of the products of decomjiosition of nitrogenous organic matter, perhaps nowhere more observable than in stables, where it is plainly perceptible to the sense of smell, it is hardly necessary to point out that its sources are various and innumerable. NITROGEN ACIDS. Nitrous and nitric acids are also present in small traces, due in part to the union of atmospheric oxygen and nitrogen through the agency of electrical discharges, and in part to the action of ozone on ammonia. Nitric acid is found in comparative abundance in buildings lighted by means of the arc light, but it is not ])robable that the amount present is of sanitary importance. AQUEOUS VAPOR. 231 AQUEOUS VAPOR. Aqueous vapor is a normal constituent which occurs in variable amounts, influenced by a number of natural conditions, the chief of which is the temperature. It is an invisible gas, lighter than air and very unequally diifused. Its sources are numerous ; some comes from the evaporation of water, some from soil moisture, some from the lungs and skin of animals and man, some from the leaves of growing plants, some from combustion. Indoors, a considerable amount is com- municated to the air through the combustion of illuminants. According to Professor Foster, an adult man gives off, under ordi- narily favorable conditions, about 4 pounds of watery vapor from the skin and lungs during twenty-four hours ; 2^ pounds by the skin, and the remainder (Pettenkofer and Voit say 10 ounces) by the lungs. An adult healthy tree of fair size gives oiF an amount which is enormous in comparison. The amount of water exhaled by plants has been esti- mated by Hellriegel to vary from 250 to 400 times the weight of the dry organic matter formed during the same time, which means that during the growth of each ton of green grass or leaves of any kind, there have been exhaled therefrom many tons of water, and that in the production of each pound of dry matter, an average of 325 pounds of water has been discharged. The evaporation of water from foliage has, among other important functions, that of keeping the temperature be- low the point where the vital processes would be interfered with. The amount of aqueous vapor which a volume of air will absorb and retain depends on the temperature. For each degree of temperature, a volume of air can take up a definite amount of vapor, and no more; and when it has taken up this amount it is said to be " saturated." The higher the temperature, the greater the amount it can hold ; and hence when a volume of air completely saturated is subjected to a change in temperature, one of two things will occur : if the temperature is increased, it can take up more vapor, and hence is no longer saturated ; if it is diminished, the aqueous vapor is in excess of the amount re- quired for saturation at the new temperature, and the excess will be condensed and precipitated as moisture. At 0° C, a volume of air takes up yi-g- of its weight of aqueous vapor ; at 1 5 °, it takes up twice as much; at 30°, four times, at 45°, eight times, and at 60°, sixteen times as much. Thus it appears that, with each increase of 15° C. in temperature, the capacity for aqueous vapor is doubled. At 15° C. (59° F.), a cubic foot of air will hold nearly 6 grains of water vapor ; at 30° C (86° F.) it will hold twice as much. Evaporation cannot go on when the surrounding air is saturated ; therefore, the presence of a body of water will add nothing to a satu- rated atmosphere. But plants and animals can continue to give ofP the vapor to an already saturated atmosphere, which, however, con- denses and deposits the excess at once, perhaps on the very surface where it is originated, as on the leaf of a tree or on the skin of man. The difference between eyaporation and transpiration, which is the 232 AIR. proper term for the giving off of vapor by animals and plants, is that the one is merely physical, while the other is a vital process dne to the action of living cells. The rate of elimination of water by the body in a state of rest de- pends upon the amount of humidity present in the air. Determina- tions bv Rubner and von Lewaschew ' demonstrated the great influ- ence of humidity in this particular. At 15° C. in moist air, the daily elimination fell to 216 grams, while in dry air at the same temperature it rose to 871. The rate rises with the temperature in both moist and dry air, and the more promptly, the greater the dryness. The outer air contains commonly from 60 to 75 per cent, of the amount necessary for saturation. In some places noted for the dryness of the air, the amount is much below ; in others, where the opposite is the case, it is above. Relative humidity is the degree of approach to saturation at any given temperature. Thus, " relative humidity 80 " means that at the observed temperature, the air holds but 80 per cent, of the amount which it can take up. Absolute humidity is the actual weight of moist- ure in a given air space. Aqueous vapor exerts a most important influence. By day, it ab- sorbs part of the sun's heat and tempers it ; by night, it acts as a pro- tecting blanket to the earth by preventing too great loss of heat by radiation. At night, the earth gives up j)art of the heat which it has absorbed during the day ; and when the air is very dry and the sky very clear, the temperature falls nuich more than when there is more vapor present to prevent loss by radiation. In the Sahara, after the hottest days, the nights are generally very cool, the temperature fall- ing sometimes 30 to 40 degrees C. in a few hours. At high altitudes also, Avhere the blanket of vapor is thin, the fall in temperature at night is very marked. Absence of aqueous vapor permits the cooling process to begin as soon as the sun gets low, and ice may form in a few hours where, during the day, the sun's heat had been intolerable. This is seen in the great deserts and at high altitudes. It is noticed commonly that the first frosts of autumn and those which come occasionally in the middle and later parts of spring occur only on very clear nights with low humidity. An amount of watery vapor apjH-oaching saturation gives I'ise to dis- comfort, whether the temperature be high or low. The "sticky" days of summer and the " raw " ones of winter owe their disagreeableness to their high relative humidity. In a hot saturated atmosphere, while transpiration can proceed, evaporation cannot, and hence the cooling influence of evaporation is missing. The sweat stays on the skin in the liquid form instead of passing into the air as a vapor, and the word " sticky " becomes singularly api)ropriate. On the other hand, with low humidity and high temperature, the sweat does not condense and remain on the skin, but passes into the air, and transpiration is not impeded in the lungs. Hence the great bearability of dry heat as * Archiv fiir Hygiene, XXIX., p. 1. DUST AND MICRO-ORGANISMS. 233 compared with moist. Saturation at low temperature has as great, if not greater, influence on bodily comfort. It does not follow that since one feels the heat more acutely with high relative humidity, this condi- tion will enable one to withstand the opposite discomfort of cold. Indeed, the reverse is true. At low temperatures, saturated air causes a greater withdrawal of heat than dry air, and intensifies the sensation of cold ; for moist air is a much better heat conductor. Cold dry air is much more comfortable than air some degrees warmer but materially moist. In the very cold climate of eastern Siberia, the air is so dry that 50° to 60° below zero F. is no hardship, provided one wears com- pletely dry clothing, while with moist clothing one would perish in a very short time. Some parts of Siberia are both cold and damp, and hence uninhabitable. Atmospheric moisture has, therefore, directly opposite effects ; it intensifies the effects of heat and also those of cold. DUST AND MICRO-ORGANISMS. Another normal constituent of the atmosphere — one of enormous importance — is dust ; normal, because it is everywhere in the atmos- phere, and because a perfectly dustless air is an artificial product obtained only with the observance of great care. The individual particles are very small, but at the same time very variable in size, ranging from those plainly discernible to the naked eye, to those of extreme minute- ness. Dust is organic and mineral, and has its origin in countless processes. It includes particles of animal matter, vegetable substances of every kind including bacteria and moulds, sea salt, matters swept from the soil by the action of winds, those discharged by volcanoes,^ others from manufacturing establishments, from chimneys, and from the millions of meteorites which daily fall from interplanetary space. The ordi- nary combustion of illuminating gas yields millions and millions of particles of carbon for every individual cubic foot. Organic dust exists only in the lower strata of the atmosphere, but that of mineral origin is everywhere. Micro-organisms are very abundant in the air of inhabited rooms, and in general in that of towns and cities, less abundant in the country, and least at great heights and at sea. Experiments have shown that at an elevation above 6,300 feet the air is free from them. Pasteur exposed a large number of flasks of broth at an altitude of 6,000 feet, and obtained a growth in but one, Tyndall exposed 27 flasks at 8,000 feet, and got no growth whatever. Dr. Fisher^ has shown that on the ocean, 120 miles from land, the air is usually free from organisms, and that at lesser distances — 90 miles, for example — it contains but few. The air of cities contains thousands in every cubic meter, against ^ After the great eruption in Java in 1883, a haze of extremely fine particles of pumice, estimated to he from seven to more than twenty miles above the earth, was visible in all parts of the world for several months. ^ Zeitschrift fiir Hygiene, I., p. 410. 234 AIR. Fig. 8. less than a hundred in the same volume of country air. It has been calculated that, in densely populated places, such as London and Man- chester, an individual inhales in the course of an hour upward of 4,000,000 of germs and spores. But this figure is enormously in excess of the figure given by Fliigge,' who estimates that in seventy years a man may inhale 25,000,000 bacteria, Avhich, he says, is about wliat one swallows in 25 cc. of ordinary milk. The number of bacteria in air is influenced very greatly by dr}'^ winds and aqueous vapor. The former, sweeping them up from the surface, increases their number ; the latter, by condensing on them and on the dust particles to which they adhere, causes them to fall to the ground. They are washed out of the air by rain, and are killed by long exposure to bright sunshine. Moulds, on the other hand, have been observed by Miquel to increase rapidly after a rainstorm, and to be much less affected by winds. The average number of organisms found at Montsouris in an inves- tigation which lasted six years was 455 per cubic meter. The lowest results were observed in Februar}' and the highest in July. During the same period, the number in the air at the center of Paris was 3,910 ; the smallest figures were yielded in Januar>' and the highest in May. All organisms are less numerous in the air at night, since then there is less mechanical disturbance of the earth's surface. AVhile the number of bacteria in outdoor air may be fairly high, it should be borne in mind that the majority of them are of the harmless varieties, and that the pathogenic kinds constitute only an infinitesimal pro- portion. Dust, as has been said, is of enormous im- portance. "Without it there would be no rain, no fog, no clouds ; the air would be satu- rated with moisture, and every object would be continually wet. Dust is largely hygroscopic, and, there- fore, attracts the watery vapor of the atmos- phere, thus becoming the nucleus for a drop of rain or particle of mist. AVere it not for its presence in the air, the aqueous vapor would condense without rain on every tree and plant, every ^ rock, every dwelling, every living creature, and, in short, on every object to which air has access. That atmospheric dust is necessary for the production of rain and fog, may be demonstrated very simply by condensing moisture from a saturated atmosphere through lowering of the temperature, and noting ' Grundiiss der llvgiene, 1897. Apparatus for demonstrating the relation of dust to rain and fog. CABBOX 3WyOXIDE, ETC. 235 what occurs when dust is present or absent. For this purpose, a simple apparatus such as is shown in Fig. 8 is all that is required. This con- sists of a large flask fitted with a rubber stopper, through which pass two pieces of glass tubing, to the free ends of which pieces of rubber tubing with pmchcocks are attached. The glass tubes project beyond the shoulder into the body of the flask. If we pour into the flask an amount of water rather more than sufficient to fill the neck when the flask is inverted with the stopper in position, we have the conditions necessary for complete saturation of the confined air with watery vapor. If now we withdraw by suction through one of the rubber tubes a small amount of the contained air, the temperature falls at once ; and inasmuch as the air within is already saturated, and since the lowering of the temperature of a saturated atmosphere is accompanied by con- densation of part of its moisture, such a condensation occurs within the flask, and is manifested by the formation of a distinct haze which fills the whole air space. If next we restore the original pressure by read- mitting sufficient air to abolish the partial vacuiun, the mist disappears instantly. The production and dissipation of the mist cloud may be repeated indefinitely so long as nothing is done to remove the dust from the air ; but if we \ya>h the air thoroughly by shaking the flask vigor- ously for a few minutes, and then repeat the experiment, no visible mist is produced. CARBON MONOXIDE, ETC. Other matters found in air include, under certain conditions, traces of sulphuretted hydrogen, sulphurous, sulphuric, and hydrochloric acids, carbon disulphide from rubber factories, marsh gas, carbon monoxide from illiuninating gas, fiimes of zinc, arsenic, and phosphorus, organic vapors from offensive trades, and other gaseous and solid matters too numerous to mention. The most important of these is carbon monoxide, a very powerful poison, often present in the air of inhabited rooms from lealdng gas pipes, imperfect combustion of illuminating gas, and defects in heating apparatus fed with coal. It is yielded in great abundance by burning charcoal, and is given oif in small amomits from stoves of cast iron, which material in a red-hot condition absorbs it in considerable amounts from burning coal. This was noticed first by Dr. Garret,^ of Chamb^ry, who described an outbreak of sickness traced by him to this cause. Later, this property of cast iron was established beyond a doubt by others. Another by no means insignificant source is burning tobacco, 1 gram of which, according to Grehant,^ yields 82 cc. of the gas. Its presence in the air of rooms in which smoking is carried on was illus- trated by Kunkel," in 1888, before a society of scientists, by exposing a small amount of blood solution to two puffs of tobacco smoke, and •demonstrating the absoqjtion of the gas by means of the spectroscope. ^ Comptes rendus, 1865, p. 793. ^ Annales d'Hygiene publique, 1879, p. 115. ■^ Sitzungsbericht der piiYsikalisch-medicinische Gesellschaft zu Wurzburg, 1888, p. 89. 236 AIR. The most important source of all is illuminating gas, which contains it in varying amounts, according to its mode of manufacture. Under ordinary conditions, the leakage of gas from the mains into the soil and thence into the atmosphere is enormous. Pettenkofer ^ reckoned that in badly jointed systems at least a fifth of the annual output is lost in the ground, and Wasserfuhr'" has calculated the annual loss in Paris due to leaks as 1 5,000,000 cubic meters. Leakage occurs from im- perfect joints, faulty cocks, and corroded iron pipes. Besides that due to leakage, we have to reckon with that due to imperfect combustion. While an Argand or other burner acting normally gives oif no trace of carbon monoxide, a certain proportion of the gas will escape oxida- tion and mingle with the air of the room together with other impuri- ties, if the gas supply is not properly regulated. The use of gas stoves is responsible for more or less contamination due to imjierfect combustion, for wlien a cold object is j^ut into the flame, the latter is cooled, and part of its carbon monoxide is given off as such. Imper- fect combustion of kerosene is still another source which should not be overlooked, for a smoking lamp exerts a very decided influence on the respirability of the air of a room, aside from the discomfort caused by the particles of soot. Less than 0.25 per cent, by volume in the air will cause poisoning, and but 1 per cent, is rapidly fatal to animal life, owing to the fact that it unites very readily with the haemoglobin of the blood cor])uscles, forming a stable c^hemical compouiul, carboxyhjemoglobin, wliich will neither take up and carry oxygen to the tissues nor promote the elim- ination of carbon dioxide. As a consequence, asphyxia occurs. In fatal cases of poisoning, carbon monoxide produces a ra])id par- enchymatous degeneration of the liver, kidneys, spleen, and heart. Carbon monoxide has been proved by L. de Saint jNlartin ^ to be present in minute amounts in the blood of animals living in cities. Nicloux * has gone farther, and demonstrated its existence in that of animals in the country, and, indeed, in about the same amounts (0.1 G volume per cent.). Nicloux finds by ex[)erim('nt that it is not derived from the air, but is develojied directly in the system, and that its amount is diminished by bringing about slight asphyxiation. Potaiu and Drouin '"' have shown that, at ordinary temperatures, it is oxidized gradually to carbon dioxide. Contamination of the air of dwellings with gas from leaking street mains is (piite common, and f^ital results are not infrequent, the gas travelling through the soil for considerable distances and being drawn up through cellars by the force of aspiration brought into play by the diiierence l)etween internal and external temperatures. Many cases of fatal poisoning have been recorded in which the gas was aspirated through the soil for more than a hundred feet. Such accidents are ^ Ueber die Vci'siftinic; iiiit Leuchtiias. Nonl unci Sud, January, 1884. ■^Deutsche Vierk-ljalirsscluift t'iir r.rtentliche Gesundheitspflege," XVIL, 1885, p. 309. " f'omi)tes rendus, CXXVI., p. 103(5. * Hrnk'ni, ex XV I., p,.. 1526, 1595. * Ibidem, CXXVl., p. H38. "SEWER GAS." 237 naturally more likely to occur in streets which, being well paved, present an obstacle to the escape of the gas upward. The odorous constituents of the gas serve a very useful purpose in pointing out the •danger, but sometimes they are held back by the earth and cannot per- form that office. Professor WoliFhiigel reported, at the Sixth Congress ■of Hvgiene, a case of poisoning by gas which had thus been robbed of its odor by the absorptive power of the soil. Water gas has much less ■odor than ordinary coal gas, and this fact, together with its much greater content of the poison, has led, as was predicted, to a notable increase in the number of fatalities due to gas leaks. Since its introduction, a very material increase has obtained in Xew York, Baltimore, Brooklyn, and Boston. "SEWER GAS." Another impurity is what commonly but improperly is called " sewer gas." This is simply sewer air which may be more or less foul by reason of containing the emanations of sewage matters. Its chemical composition depends upon the extent to which the gases of decomposi- tion are generated, and upon the rate of ventilation. It may be almost as pure as the outside air ; it may be as rich in carbon dioxide as the air of badly ventilated rooms ; and it may be much worse. From 10 to 30 volumes of CO., in 10,000 are found quite commonly. Dr. W. J. Eussell found as high as 51 volumes in 10,000 in the air of one ■of the London sewers, and Letheby as high as 53.2 in that of another, while in an unventilated sewer in Paris as high as 340 volumes have been reported. Sulphuretted hydrogen and ammonium sulphide are ordinarily present in small amounts or mere traces, and may be wholly absent ; but in old unventilated sewers, they may be present in notable amounts. The highest recorded, 299 volumes in 10,000, was fomid by Parent-Duchatelet in an old choked sewer in Paris. Marsh gas, ammonia and compound ammonias, and other gaseous products of de- composition of organic matter, may be present in variable amounts, according to circumstances. Sewer ah- contains micro-organisms and animal and vegetable debris, just as does the outer air ; but, as a matter of fact, the number of bacteria is invariably small, and they are often wholly absent. This was shown first in 1883 by Micpiel, whose results have been corrob- orated by those of a number of other investigators, including Carnelly -and Haldane, Laws and Andrews, and Percy Frankland. The first mentioned conducted a most elaborate chemical and bacteriological ex- amination of sewer air, and proved that from both points of view it compares favorably with the air of schools and small dwellmgs, and that bacteriologically it is, indeed, far superior. It contains fewer organisms than the air of the streets above or of any kind of dwelling, and such as are present come entirely or chiefly from the outer air, and not from the sewage. Laws and Andrews arrived at the same conclusions after a similar Tesearch. In each sample of sewage examined, B. coli communis was 238 AIR. found in numbers varying from 20,000 to 200,000 per cc, and closely allied species in even greater abundance ; but neither the one nor any of the others was found in the many samples of air examined. Thev found, farther, that the number of organisms existmg in sewer air depends entirely upon the number present in the outside air in the immediate vicinity, and that while sewage bacteria are largely of the liquefying vai'ieties, such are practically abseut in the air. The chief importance of "sewer gas " lies not in its power to jiro- duce disease, but in its capacity for being the vehicle for odors which make the air disagreeable, but not necessarily dangerous to health, except that a])petite and digestion, and hence general nutrition, may be interfered with. As a matter of fact, sewer air has served for a long time as a most convenient scapegoat in investigations of the cause and spread of out- breaks of typhoid fever and other infectious diseases, and as a most useful aid in explaining obscure questions of various sorts. !Many be- lievers in the sewer-air theory of dissemination of typhoid fever hold that the coarser dust particles carry the germs on their surface, and may be blown about through considerable distances before the organisms lose their vitality ; but the great objection to this explanation is that in sewers and cesspools the tyj)hoid l)acillus is destroyed speedily by other organisms, and that, even though it be present in an active state in liquid sewage, it is extremely unlikely that it will be released therefrom into the air. No ordinary stirring up of the water will throw the germs into the air ; although, according to the researches of Fraukland,^ the bursting of gas bul)bles generated by decomposition will throw into the air traces of chemical salts which have been mixed and di.^solved in the sewage, and in the same way may throw out bacteria as well. But it has been shown by Xageli that bacteria cannot be given off from moist surfjices. Another explanation, offered by Dr. C. R. C. Tichbourne,^ is that the disease germs are scattered into the air by the fermenting sewage, and carried by a mist formed when the warm sewer air, saturated with moisture, meets the colder external air at the points where ventilating outlets are placed. Then each minute droplet of mist, carrying one or more microbes, is trans])(»rted through longer or "shorter distances in the air, perhaps into dwellings, and eventually, by the influence of the heat of the sun or by other natural agency, becomes dissipated as vapor, and leaves the organisms suspended in the atmosphere. The majority and the best of the German investigators, as Fliigge, Rubner, Gartner, Soyka, Prausnitz, and others, maintain that sewer air and sewer gases are incapable of conveying the germs of typhoid fever and other infective diseases. It is true that some of the gases given off in the putrefactive processes which go on in sewers are more or less poisonous, but whether they are capable of produc- ^ Proceed in jfs of the Royal Society, 1S79. * Dublin Journal of Medical Sciences, July, 1897. "SEWER GAS." 239 ing injurious eiFects depends very much on the amount inhaled and on the degree of concentration. In anv event, thev are certainlv in- capable of producing any infective disease in the absence of the specific germ. Ill anv well-constructed and properly ventilated sewer, no great amount of putrefaction will go on, smce the sewage matters soon pass on and are discharged ; consequently not much gas will be evolved, and, with proper ventilation, whatever is evolved, is soon dissij^ated in the outer air. Offensive gases and odors are much more likely to be given off by unclean unventilated house-plumbing than Ijy ^vell-l)uilt sewers. It is asserted co mm only that the inhalation of small amounts of this air will produce headache, anemia, loss of appetite, sore throat, albu- minuria, diarrhcea, and other symptoms, and that it may be the exciting or auxiliary cause of tj-phoid fever, measles, diphtheria, scarlet fever, dysentery, and other infective diseases. But iu the cases which are accepted as proving the causal relation, inference has taken the place of proof, no other means of infection being ascertainable. In not a single case has the supposed relation been demonstrated bacteriologically. In answer to the well-known stubborn fact that the workmen em- ploved in all the large systems of sewerage — men whose occtipation involves the daily and constant inhalation not of traces, but of large volumes, of sewer air — are as a class unusually healthy and strong, with a high mean age at death and a low death-rate, it is asserted that they become immunized by daily contact, and thus escape. If we accept this theory, however, we should go farther, and say that large doses are a benefit in that they confer immunity, and that, therefore, all precau- tions against the admission of sewer air to the ah" of dwellings are mis- directed, and should be abandoned. The air of properly constructed sewers is in constant motion, brought about by differences in temperature and mechanically through influx of sewage. During the colder months, the temperature within the sewer is higher than that of the air above, and it is influenced materially by the fact that the entering sewage is largely Avann ; therefore, sewer air tends to rise and escape through the openings in the man-hole covers. During the warm season, the natural interchange is much lessened, since then the conditions as to temperature are reversed. At night, however, at all seasons, the temperature of the air of the sewer is higher than that of the atmosphere above, and thus ventilation goes on by natural laws the year round. ]Much air is displaced by the entering sewage ; in fact, disregarding the effect produced by the warmth of the sewage, for ever}- cubic foot of sewage which enters, a cubic foot of air is forced out, and as the sewage is discharcred, more air enters to take its place. Owing to the prevailing belief in the noxious character of sewer air, it was formerly the custom to place baskets of charcoal in the out- let shafts of the man-holes, but as this material loses its absorptive property with access of moisture, the plan was abandoned. It was also 240 AIB. regarded as an advantage to connect the sewer with chimneys, which act as ventilators, but in the light of farther knowledge and because of excessive action, that method of ventilation fell into disuse. ORGANIC MATTERS. Among other impurities given off to the air, the organic matters from the processes of the body are, in a way, of considerable impor- tance. These include particles of epithelium, the constituents of sweat (butyric, capric, capronic, and caprylic acids, lactate, butyrate, and other salts of ammonium), and volatile matters from foul mouths, decaying teeth, and the digestive tract, and excrementitious matters deposited on unclean clothing. In addition to these, it has been asserted that other matters of a poisonous character are given off in the process of respiration, which matters will be referred to later on in the discussion of the effects of impure air on health. That the air of inhabited confined spaces may contain organic animal matter, is appar- ent to the senses when one enters such an atmosphere from one not thus contaminated. Effects of Vitiated Air. The effects of foul air on the system are of great importance, and vars' in degree within veiy wide limits. For proper aeration of the blood, it is necessary that the oxygen of the air shall be present in the normal proportion in the free state, and not in chemical union with carbon as a waste product. Farther, it is necessary for the proper ex- cretion t)f the carbon dioxide of the blood that the difference in the tension of that gas in the air and of that in the blood shall be as wide as possible ; that is to say, the less the amount of carbon dioxide in the inspired air, the greater the facility with which the blood can dis- engage that which it carries to the lungs. Any interference Avith this most important function of the body must have an uijurious effect on the general health, and it is accepted generally that impurity of the air is, Avitliout doubt, the most important of the predisposing causes of disease. It is well known that, other conditions being equal, in j)roportion as a people are drawn to employments indoors, the disease-rate and death- rate are increased. This is particularly true as regards phthisis, which is preeminently associated with overcrowding. Overcrowding means the association of two or more people in a space so confined as to preclude the admission of a constant sup]ily of fresh air sufficient in amount to maintain a proper dilution of their excretory ]iroducts and a normal su|ij)ly of free oxygen. It was recog- nized long ago as a most important factor in tlie production of a high death-rate among occuj)ants of crowded jails, barracks, and liosi)itals ; and experience lias demonstrated repeatedly that increase in space allowance is followed always by decrease in sickness- and death-rates. At one time, for example, the English army averaged 11.9 deaths per EFFECTS OF Y IT Li TED AIR. 241 1,000 men annually, from phthisis alone; more efficient ban'ack ven- tilation and increase of average air space caused immediate improve- ment, and the phthisis-rate fell gradually to 1.2 per 1,000. The same general result has been observed in the armies of France, Russia, Germany, and Belgium. AVhat is true of overcrowding applies not alone to human beings, but to animals as well, and it is a well-known fact that crowded stal^les show high mortality among cows and horses. It has such a remark- able influence on egg production and gro\\i:h of fowls that practical poultrvmen are exceedingly careful on this point. The immediate effects of inhalation of impure air are discomfort and oppression, which may amount to headache, dizziness, faintness, and even nausea. Continued exposure is likely to bring about a gradual impairment of health, shown by pallor, languor, anaemia, skin troubles, loss of appetite, and diminished power of resistance to the exciting causes of disease, and this is especially true of those whose daily work is carried on in crowded spaces. It is customary to cite as extreme cases of overcrowding and its effects, the Black Hole of Calcutta, the ship Londonderry, and the prison at Austerlitz ; but the conditions that obtained in each of these instances were most unusual, and the cases are of historical rather than sanitary interest, since the confining of a number of persons in a space from which air is practically excluded can have but one outcome. The Black Hole of Calcutta is the name applied to the military prison of Fort ATilliam, where, in June, 1756, Surajah Dowlah con- fined 146 persons over night in a space of less than 5,900 cul^ic feet, A\-ith two small windows in one side. Within an hoiu', all broke out in a profuse sweat, and were tortured with thirst and difficult breathing ; in three and a half hours, a majority were delirious, and when the place was opened in the morning, 123 of the prisoners were found dead. In the case of the Londonderry, which, in December, 1848, left Sligo for Liverpool and ran into a storm. 2O0 steerage passengers were con- fined over night in a space 18 by 11 by 7 feet, witli no means of ven- tilation. In the morning, when they were released, it was found that over 70 had expired. In the other extreme case, that at Austerlitz, 300 captured soldiers Tvere confined in a small cellar, and within a few hours all but 40 were dead. To what one or more conditions of impure air are the ordinary effects due? TTe have seen that COo is in itself not an active poison, and that its action is tn interfere with the proper oxygenation of the blood within the lungs. The aqueous vapor of respiration and from the skin, and that produced in the combustion of illuminating material, constitutes an important part of a vitiated atmosphere, and is respon- sible for at least a part of the discomfort produced ; but it is also true that a deficiency in wateiy vapor in the air of well-ventilated rooms has equal or greater disadvantages, as will appear in the consideration of Ventilation. 242 AIR. Concernino; the effect of usual amounts of ordinarv dust in in- habited rooms, there is little to be said. The micro-organisms, most of which are non-pathogenic, vary in number with efficiency of ven- tilation. In pure air, the bacteria and moulds approximate each other in number ; but in vitiated air, the l)acteria increase in number, Avhile the moulds are much less affected. The experiments of Carnelly, Haldane, and Anderson showed a progressive increase in both bacteria and moulds with diminished ventilation. Thus, Character of air space. Number organisms in 10 L. air. Ratio of moulds to Moulds. Bacteria. bacteria. External air 4-roonied houses 2-roomed liouses l-roomed houses 2 4 22 12 6 85 430 580 1:3 1:21 1:20 1:48 The increase in bacteria is not due to respiration, though a diminu- tion in their number might be thus explained ; for the great majority of inhaled bacteria are hltered out by the nose, and the ex])ired air is almost completely free from germs, although they may be thrown out in the act of coughing or sneezing. Investigation thus far has not proved that the bacteria of infection are commonly introduced into the system through the medium of re- spired air. As has been mentioned, it is held by many that the effects of vitiated air are not due to carbon dioxide, but to the organic matters and aqueous vapor given off by the lungs and skin, and that these are estimated conveniently by determining the amount of carbon dioxide with which they are discharged. It is said also that, while considerable carbon dioxide escapes even under the most imperfect system of ventilation, the organic matters and watery vapor do not so readily pass out, but are deposited on walls, furniture, hangings, and clothing, where they putrefy and become offensive. As proof of this, is cited the fact that a room in which a per.'^on has slept without adequate ventilation has an un})lea.'^ant smell in the morning, and tliat this persists even after prolonged airing. Brown-S^quard and d'Ar.>^onval, in 1888, obtained from condensa- tion of the aqueous vapor of men and animals a liquid which, injected into rabbits, caused death with greater or less rapidity, according to the size of the dose. They l)elieved the toxic element to be of the nature of a volatile alkaloid, and that it was exhaled dissolved in the aqueous vapor of the breath. In the sjme year, Wurtz, reporting a similar research, claimed to have found a toxic substance. Merkel,' in 1892, claimed to have obtained jwsitive results, and con- cluded that respired air from ])ersons in health contams a minute quan- tity of a volatile organic base, which is poisonous when free, but innoc- • Archiv fiir Hvgiene, XV., p. 1. EFFECTS OF VITIATED AIR. 243 uous after contact with an acid. Dr. Sivierato ^ collected the aqueous vapor of the breath of persons suffering from diseases of respiration, both with and without fever, of persons with no respiratory disease, but wdth fever, and of persons in health, and injected it into rabbits. That from those with respiratory diseases produced fever and diminished reflexes lasting three to six days ; that from cases of fever with no res- piratory disease caused little or no disturbance ; and that from persons in health produced no results whatever. Formanek - concluded, after much study, that no poisonous substance originates in the lungs ; that the ammonia sometimes found is not a product of metabolism, but of decomposition in the mouth cavity (cari- ous teeth, etc.) and in the trachea and lungs after tracheotomy, and in pulmonary tuberculosis ; that, in the experiments which led to the theory of an unknown alkaloid, ammonia was used, and might have caused the observed efPects ; and that the results of overcrowding can- not be due to any one cause. Many other experimenters, French, German, Italian, American, and English, working along the same lines, but with extra precautions to exclude matters from the nose ,and mouth, have failed to obtain toxic effects from the condensed vapor ; others have demonstrated that the lungs exhale no organic matter except in minute amounts, and that these have no poisonous mfl^uence. Arloing, however, has pursued the subject farther, so as to mclude the sweat constituents among the deleterious agents of vitiated air. He soaked in distilled water the undershirt of a man who had spent a long evening in dancing, and injected the infusion into dogs ; they became drowsy, suffered from violent diarrhoea, and shortly died. The drawers, similarly treated, gave a liquid which, injected into rabbits, produced clonic spasms follow^ed by paralysis and death. Should these results be confirmed by a more scientific research, in which the possible influ- ence of the body bacteria and other agents are eliminated, it will appear that the sweat possesses a high degree of toxicity. Until such confirmation is reported, however, the weight of evidence leads to the conclusion that the injurious action of vitiated air is due to the duni- nution of oxygen and to the increase of carbon dioxide, both of whicli factors, alone or together, interfere with the intake of oxygen and the excretion of carbon dioxide from the lungs. Yet, diminution in oxygen, which even in very crowded rooms does not proceed ^'ery far, is met by increase in the respiratory function, which, liowever, cannot increase the diflPerence between the tension of the carbon dioxide of the air and of the blood. Xot even in very imperfectly ventilated mines does the oxygen fall much below 20 per cent, by volume, and thus we see that the whole range of fluctuation in the oxygen of pure and of very foul air is but little more than 1 volume per cent. Smith and Haldane ^ have shown that in a leaden chamber containing ^ Archives Italiennes de Biologie, 1895. ^ Archivfiir Hygiene, XXXYIII. (1900), p. 1. ■' Journal of Patliologv and Bacteriology. I., 1892. 244 ATE. air which had suffered but slight diminution in oxygen, but which contained 384 parts of carbon dioxide in 10,000, two men suffered from headache immediately on entering. As a rule, vitiated air is associated with high temperature and satu- ration with aqueous vapor, which latter interferes with evaporation from the skin. Less often it is associated with low temperature, and with this condition comes an increased demand for oxygen to meet the requirements of the oxidation processes. It seems probable that where the carlion dioxide is not present in any great excess, and the oxygen is not markedly deficient, the conclusion arrived at by Drs. AVeir Mitchell, Billings, and Bergey is true ; namely, that the discomfort suffered is due largely and chiefly to heat and dis- agreeable odors arising from the occupants in various ways : from bad breath, unclean skin, unclean clothes, sweat, and gases from the bowels. Such Jiiay induce very disagreeable sensations, amounting even to nausea, in those who are not habituated to such influences ; but, on the other hand, those who are accustomed to such air notice no discomfort. Disagreeable smells do not act directly as a cause of specific disease, but appear to have an influence on the ajipetite, and hence on the gen- eral well-being of persons not accustomed to them. Much is due also to the imagination ; a disagreeable smell from a source known to be clean (chemicals, for instance) has not ordinarily as much influence as another of equally ofl'ensive character sujiposed to be from filth. It seems prolial)le also that there is much to learn concerning the real effects of disagreeable smells, and that they may be more extensive than we now commonly believe ; but in order to determine this, we shall need methods which will reveal the nature of the odoriferous substances and make their isolation possible. Other causes of discomfort may be sought for in the presence of traces of carbon monoxide from heating apparatus or incomplete com- bustion of illuminating gas, and in excessive diyness of the air due to furnace or steam heat. It should not be overlooked that impure air may affect the vitality and bactericidal power of the cells of the air-jiassages and of the ali- mentary tract, and thus lessen the power to resist the action of infective material. The Air as a Carrier of Infection. On the agency of air in spreading infectious matter, much has been said and written, and much careful research has been conducted, but the conclusions reached are by no means in agreement or conclusive. It is conceded generally thatVpiithogcnic organisms in the air are ad- herent to particles of dust of various kinds, and that their retention of virulence depends upon the amount of hygroscopic moisture with which they are associated. The conditions favorable to their continuance as living organisms are naturally more likely to obtain in indoor air, with imperfect ventilation, than in the outer air, where they are diluted and blown about and exposed to the disinfectant action of the dii'cct THE AIM AS A CARRIER OF INFECTION. 245 rays of the sun. Indoors or outdoors, the more they are protected by hygroscopic dust particles, the longer they will retain the moisture which is essential to their viability. It appears, too, that, conditions being equal, certain micro-organisms retain vitality longer than others, some being but slightly, others very tenacious of life. With regard to the transmission of pulmonary tuberculosis through the air, it should be said that while there can be no doubt that this dis- ease is connected preeminently with overcrowding and vitiated air, there is a very decided difference of opinion as to the method of con- veyance, some contending that dust, and others that tuberculous material, thrown into the air in coughing, speaking, and sneezing, is the vehicle. Buchner has found B. tuberculosis in an active state in the dust of a room a year after the death of its occupant from the disease. G. Cornet^ demonstrated its presence in more than a third of 147 samples of dust collected in hospitals and other public institutions, and in private houses inhabited by phthisical persons, and succeeded later in producing the disease in 46 out of 48 guinea-pigs exposed to air containing dust from dried tuberculous sputum. Some of the animals were placed 8 inches from a glass vessel containing dried pulverized sputum from an advanced case ; others were placed on shelves 8 to 28 inches from the floor of a room, on the carpet of which, sputum, mixed with dust, had been spread and dried and, at the end of two days, stirred up by sweeping ; others were allowed to stay in the room without disturbance of the dust. Klein obtained positive results with guinea-pigs placed in the venti- lating shaft of a consumptives' hospital ; but Heron ^ obtained but 2,7 per cent, of positive results in 74 guinea-pigs inoculated with dust from the ventilating shaft of the London Hospital for Diseases of the Chest ; and Kirchner ^ got but 1 positive result out of 1 6 pigs inoculated with the dust from a military hospital. Fliigge, on the other hand, was wholly unsuccessful in inducing the disease in guinea-pigs exposed to such dust ; and concluded that the transmission from one person to another is chiefly by means of the finest droplets thrown into the air in speaking, coughing, and sneezing. From later experiments, conducted under his supervision by Laschtschenko, Heymann, Sticher, and Benincle,^ he concluded that in rooms in which tuberculous sputum is dried on the floor or other places, and where the air is filled with coarse dust through dry cleaning and air currents, or, as in railway cars, by continual mechanical jarring, infection may arise ; and that, under these conditions, long-continued exposure offers a certain degree of probability of infection. Therefore, dry cleaning is to be avoided in rooms in which consumptives are employed with others, and the rooms should not be occupied so long as the air is perceptibly dusty. The great pos- sibility of infection through matters thrown off in coughing and sneez- ] Zeitschrift fur Hygeine, V., p. 191. ■' The Lancet, January 6, 1894. * Zeitschrift fiir Hygiene und Infectionskrankheiten, XIX., p. 153. * Ibidem, XXX., p.l07. 246 ATR. ing is insisted upon as of paramount importance. This danoer is to be ])revented by requiring the person coughing to hold a handkerchief or the hand before the month during tlie act, and by the avoidance on the part of others of approaching within a meter. Answering Fliigge, Cornet ' contends that the number of bacilli thrown into the air during the act of coughing must be extremely small. He caused IS consum])tives to hold dishes before the month while coughing, and obtained 2 positive results therefrom on inoculation into guinea-pigs ; repeating the test with 15 others, he got none. Zielgen ^ also disagrees with Fliigge as to the possibility of trans- mission of germs in general through the air, and believes that it is assisted much by evaporation of water from surfaces upon which they are deposited. He obtained positive results from inoculation of drop- lets condensed on the upper surface of a vessel in which a culture of B. pi/oci/a)ieu,s had been incubated. Experiments conducted by Koniger'* confirm Fliigge in his estimate of the danger of transmission by droplets. In order to give the ex- pelled droplets a character which wt»uld admit of their being traced, he rinsed his mouth with liquid rich in B. prodiglosus or B. mycoides, or with very dilute caustic soda, and, in order to trace them, he exposed Petri dishes and glass plates coated with phenolphthalein, which agent, turning pink in contact with an alkali, would show not only the number of droplets, l)ut their size as well. It was found that no drop- lets are thrown out in ordinary exhalation nor in vowel formation, but with consonants, as t, k, and y>, the number is very great, and is largely dependent ujion the amount of force with which tlie air under ])ressure in the mouth cavity is released in their formation, and, therefore, upon the manner of pronouncing. Loudness and ra})idity of speech have but little influence ; Avhispering may, indeed, under some conditions, cause a greater nimiber of dro]>lets than loud speech. Even with sub- dued speech and a quiet atmosphei'c, it was found that the organisms expelled reached the most distant jiarts of the room, which was more than 20 feet in width, and in all directions. They were found to remain in suspension in the air not longer than an hour, and it was noticed that they fell upon the plates in grou]is, sometimes as many as 40 close together, which suggests that they fall not as dry dust particles, but that the droplets themselves, with their contained or adlierent organisms, are deposited. In coughing and sneezing, more droplets are exjielled than in speaking, and they are projected to a greater dis- tance, because of the greater force engaged. The ]>rccantions recom- mended a])ply not alone in tul)erculosi\«3_l^it also in diphtheria, whoop- ing-cough, and other diseases in which the respective specific organisms are found in the air- passages. Hutchison * found that bacteria, sprayed in minute droplets upon ' Berliner klinische AVocIu'iiscluit't. Mav 13, 1899. ^ Tribune m^dioale, May 10. 1S99. 3 Zeitsciirit't fiir Hvffiene und Infeolionskrankheiten, XXXIV. (1900), p. 119. * Ibidem, XXXYI. (1901), p. 223. THE AIR AS A CARRIER OF INFECTION. 247 objects, perish in a short time, the main factor in their destruction being the influence of sunlight. Sprayed directly into the air, most of them were found to have become deposited within a half hour, when the air of the room was allowed to remain undisturbed, but numbers of them were kept in suspension for considerable periods by slight unavoidable air currents in the lower strata. He showed that, with favoring air currents, the suspended bacteria may be conducted through very nar- row crevices, as into closed bureau drawers, and from one room to another through keyholes and cracks. While the danger of dissemi- nating bacteria by walking over an infected floor was found to be slight, those thrown up by the elastic rebound of the boards failing to infect plates suspended 4 inches above them, ordinary sweeping was found to contaminate the atmosphere throughout its whole extent, even to the ceiling, thus confirming Fliigge's statement as to the undesira- bility of dry cleaning. Closely similar results were obtained by Kirstein,^ who concludes that ordinary air currents cannot detach living organisms from surfaces upon which they have been deposited and become di'ied, but concedes that, when the bacteria are sprayed upon fine dust particles, they may easily be borne about in the air. Yet how slight the danger of this method of infection is, so far at least as typhoid fever is concerned, is shown by the marked rapidity with which the typhoid organisms die when sent forth in the form of spray. Other non-spore-builders, sprayed into the au', retained their vitality for only a comparatively short time, because of the influence of light and air ; and he believes that the marked sensitiveness of the tubercle bacillus to the influence of light makes early destruction of this organism most probable when it is thrown into the air in minute droplets, and that thus may be ex- plained the fact that, even in consumptive wards, in which there is, without doubt, a constant discharge of bacilli into the air, attempts to detect living organisms in the dust, etc., fail, excepting in those cases in which the sputum itself has, through lack of care, become dis- seminated. Positive results of examination of droplets expelled by consumptive patients during coughing have been recorded by Curry,^ Boston,^ and others. Curry experimented with 12 patients, who coughed toward plates suspended from 1 to 3 feet distant ; he found the bacilli in the larger droplets expelled by half the subjects. Boston, observing fine droplets being ejected from the mouth of a patient with advanced disease in each act of coughing, concluded that such constant spraying at the table and elsewhere might afford an explanation why patients in the early stage of the disease did not do well in the institution where his observations were made, in which every possible attention is given to ventilation, light, and disinfection of sputum. By means of a simple device, the spray sent out by 50 patients was collected, and then sub- ^ Zeitschrift fiif Hygiene und Infectionskranklieiten, XXY. (1900), p. 123. ^ Boston Medical and Surgical Journal, October 13,1898. * Journal of the American Medical Association, Sept., 14, 1901. 248 AIR. jected to examination for the specific organism, which M'as found in 76 per cent, of the cases. The smallest numljer found in any specimen was 4, and in fully a third the bacilli were very numerous. Raveuel's ' experiments with tuberculous cows have proved that they, too, send forth the bacilli in great numbers in the act of coughing. Experiments conducted at the Adirondack Cottage Sanitarium by Dr. I. H. Hance," for the purpose of determining the degree of danger of infection when all possible sanitary measures for disinfection of sputum are enforced, support the view that dust in the air is of secondary importance, but that, where carelessness in this regard ob- tains, the danger is a real one. A complete examination was made of the group of buildings, some of which had been occupied by consump- tives for eleven years. Dust was collected from places most likely to be infected, and with it 81 guinea-pigs were inoculated. Four inocu- lated with dust from the infirmary (a liuilding where all the acutely sick are sent), and from the main building (in which are a parlor, sit- ting-room and library), died of other infections on the third to the sixth day. Five of the ten inoculated with dust from the oldest cot- tage, which was occupied by a man who had been complained of for promiscuous spitting, became tuberculous. Those inoculated with the dust from the other buildings gave negative results. During eleven years, not one of the 20 to 25 attendants employed had developed the disease. As to typhoid fever, too, opinions are at variance. Dr. John Brownlee reported before the Glasgow Philosophical Society an ex- periment proving that the specific bacilli can live in ordinary dust. Buchner is of the opinion that neither typhoid nor other fevers ciin thus be spread. Perhaps the most extensive research on the subject of transmission of this and other diseases is that conducted by Dr. Eduardo Germauo.* In his experiments, he used various kinds of dust and dirt, and from his results, he concludes that the typhoid germ is unable to withstand complete drving, and hence cannot be transmitted to man through dust sufficiently dry to be disseminated by air currents. Experiment showed that the germ can live for a long time in moist surroundings, even in an apparently dry condition, that is, when adherent to or encompassed by matters which contain a certain amount of moisture, such as cloth- ing, particles of dirt, and fecal filth. ]\Iost of the bacilli die as drying progresses, but some are more or less resistant, though not necessarily dangerous on admission to air currents, sincehtheria bacilli were present in the air which had not undergone di.-infection. The bacteriological tests were controlled by inoculation experiments with animals. AVith regtird to the possilHlity of spreading cholera germs through the agency of moving air, Dr. X. William - has reported that, while that means has been regardetl as most favorable, in actual experLinent it fails. Mixed with dry dust, the germs live but a short time, and perish more quickly when a current of air is conducted through the dust. AVhen the dust is distributed through large volumes of air, the germs die rapidly, and when the impregnated dust is let fall u}X)n a suitable culture, only a ver}- small proportion of living organisms can be found. In other words, cholera germs, adherent to dust particles floating in and moved about by air, do not retain their activity for any length <»f time nor through any considerable distance. The experiments of Honsell ' indicate that the cholera organism finds no favoring conditions for its passage into the air from its situation in privy vaults. The subject of danger of cholera infection by dust from baled rags was considered thoroughly at the Dresden Cholera Conference, and it was found impossililc then to quote a single case in which infection could be traced to this source. According to Dr. E. AV. Hope,* atmospheric dust is largely respon- sible for the spread of infontile diarrhcea in cities and large towns, where, from unavoidable causes, the air becomes more or less laden with filth. He presents evidence of the association of rainfall and its attendant cleansing of the atmosphere with diminished mortality from choleraic diarrhcea, as follows : Period. Average rainfall June to September. Conditions. v. Annual average of deaths from diarrha-a during -^ third quarter of year. 6 years 13.8 inches Average wet sum- mers. 373 14 " 10.9 " Average dry sum- mers. 573 Extreme vears. 1891" lO.O " Wettest summer. 203 1895 7.7 " Driest summer. 819 ' Gazette des Maladies infantiles, No. 10, 1899. ^ Zeitschrift fiir Hygiene und lufectionskranklieiten, XV. (1893), p. 166. ' Arl)eiten :uis dcin i)atholog-anatoniiseben Institut zu Tiibingen, 1896. « Public Health, July, 1899. EXAMINATION OF AIR. 251 Influence of Yog. Dust and moisture together in the form of fog aifect the health of large communities in a marked degree. In a still air nearly or com- pletely saturated with aqueous vapor and containing ordinary dust and smoke, a fall in temperature causes each particle of dust and soot to become the nucleus of a minute droplet of condensed moisture. These countless droplets in a state of suspension form a more or less dense blanket of fog, which impedes dispersion of the impurities given off by natural processes and as products of combustion. While ordinary country and seashore fogs are not known to exert deleterious effects, in smoky cities, like London, the case is quite different. It is a well-recognized fact that, during periods of heavy fogs in manufacturing centers, the morbidity and mortality from respiratory disease are increased very greatly, and that, as the atmosphere clears, a sharp decline follows. In London, for example, the usual death-rate from all causes has been known to become almost doubled during a fortnight of continued dense, smoky fog, and then to return to its normal figure with the advent of clear weather, the increase being due particularly to bronchitis and other affections of the respiratory tract, attributed to the irritating influence of the finely divided particles of soot and the acids which accompany them. During the prevalence of thick fogs, the air being necessarily in a stagnant condition, it has been observed that the carbon dioxide content increases progressively. During one such period following bright weather, the air of London acquired, in four days, three and a half times its normal content of this gas. The importance of smoke, both as a promoter of disease and on ac- count of its corrosive and disfiguring action on buildings, and also on account of the obstruction of light, has led to much legislation and to the exercise of inventive genius for devising means for the prevention of its discharge in objectionable amounts into the atmosphere of cities. Many patents have been granted for smoke-consuming devices, the majority of which have been found to work unsatisfactorily. The most effective invention, which gives promise of solving the problem most completely, is one which has been brought to the attention of the Department of State by Consul General Mason. ^ This process consists in distributing heated and slightly compressed air through hollow grate bars to the whole lower surface of the furnace. Not only is practically perfect combustion attained, but immense saving of expense is possible, since what are ordinarily unsalable low-grade coals can be employed to greatest advantage. Examination of Air. For all practical purposes, the examination of air may be restricted to the determination of the amounts of aqueous vapor and carbon dioxide. The essential element, oxygen, fluctuates within such very 1 Consular Keports, 1899, p. 491. 252 AIR. narrow limits that its estimation is a matter of purely scientific interest, and, moreover, the process is one which deraands a much higher clegree of manipulative skill than is possessed by those to whom the task of making sanitary examinations ordinarily falls. The chief con- stituent, nitrogen, is practically constant in amount, and its determina- tion would serve no useful purpose. Whatever is the cause of the deleterious effects of an atmosphere vitiated by respiration, whether it be carbon dioxide or tiie organic matters given off by the body, this at least is certain, that the amount of carbon dioxide serves as an index of impurity, and that the amount of aqueous vapor is of considerable sanitary importance. In special cases, it is important to look for that most dangerous contamination, carbon monoxide, which, coming even in very small amounts from leaking gas pipes and other sources, exerts a decidedly deleterious influence. In the minds of many, the test for ozone is also of importance. In addition to chemical analysis, the determination of the amount of dust and the number and varieties of micro-organisms present may be of interest and importance. Determination of Aqueous Vapor. — As has been stated above, a volume of air at a given temperature can hold a definite amount of moisture, and no more, and when this amount is present the air is said Fig. 9. Apparatus for direct dctenniiiatidn of moisture. to be saturated. The amount mIu'cIi a volume of air contains consti- tutes its absolute humidity, and the difference between this and the amount which it is possible for it to hold is known as its saturation deficiency. The ratio which its absolute humidity bears to its possible content is known as its relative humidity. Direct Determination of Moisture by Weighing. — Prepare two wide- mouthed flasks of about loO ec. cajiacity in the following manner: EXAMIXATIOy OF AIR. 253 Provide each with a tightly fitting rubber stopper with two perforations, through which are inserted two pieces of glass tubing bent at a right angle. One of these reaches to the bottom of the flask, and serves as an inlet ; the other extends only a short distance below the stopper, and serves as an outlet. Fill the flasks with small pieces of pumice which have been heated to a high temperature over a Bunsen burner, dropped while hot into concentrated sulphuric acid, removed therefrom, and quickly drained. The two flasks thus filled, and \\ix\\ stoppers tightly inserted, are then to be connected by means of a short piece of rubber tubing, the inlet of one joining the outlet of the other. They are then weighed. The flask with the free outlet ttibe is now to be connected with an aspirator, by means of which from 20 to 50 liters of air are drawn through. As the air comes in contact with the pmnice sattu'ated with sulphuric acid, its moisture is absorbed and retained. At the expiration of the aspirating process, the flasks are disconnected from the aspirator and again weighed. The increase in weight represents the amount of moisture in the volume of air used. The apparatus is shown in Fig. 9. Know- ing the temperature of the air, one can then easily determine the relative humidity by refer- ence to the table below, which shows the maxi- mum humidity possible at diiferent temperatures. TABLE OF M.lXBimi WATEK CAPACITY FOE TEX LITERS OF AIE. Fig. 10. Tempera- Corre- 1 ture centi- j sponding \ Grams, grade. : degrees V. ■. I Tempera- ture centi- grade. Corre- i sponding I Grams, degrees F. | —10 — 8 — 6 — 4 — 2 1 2 3 4 5 6 7 8 9 10 11 12 14.0 0.021 17.6 0.027 21.2 0.032 24.8 0.03S 28.4 0.044 32.0 0.049 33.8 0.052 35.6 0.056 37.4 0.060 39.2 0.064 41.0 0.068 42.8 0.073 44.6 0.077 46.4 O.OSl 48.2 0.08S 50.0 0.094 51.8 0.100 53.6 0.106 13 14 15 16 17 18 19 2a 21 22 23 24 25 26 27 28 29 30 55.4 0.113 57.2 0.120 59.0 0.128 60.8 0.136 62.6 0.145 64.4 0.151 66.2 0.162 68.0 0.172 69.8 0.182 71.6 0.193 73.4 0.204 75.2 0.215 77.0 0.229 78.8 0.242 80.6 0.256 82.4 0.270 84.2 0.286 86.0 0.301 Determination of Relative Humidity by the wet and dry Thermometer Bulbs. This instru- ment, which is known also as the psychrom- eter, consists of a pair of accurate thermom- eters on an upright support. The bulb of one is free ; that of the other is covered with a Psyehrometer. 254 AIR. layer of muslin kept moistened by means of a piece of wickiug which clips into a small vessel of water beneath. (See Fig. 10.) In a satu- rated atmosphere, no evapomtion can occur from the wet muslin ; but in one not saturated, the process goes on with varying rapidity. Evap- oration is a process which requires heat and causes a lowering of the temperature of the moist surface ; the more rapid its rate, the greater the abstraction of heat. The drier the atmosphere, the greater the rate of evaporation, and, therefore, the greater the fall in temperature. If the instrument is placed in a saturated atmosphere, the two thermom- eters will give the same readings ; but in one not saturated, the wet thermometer will fall gradually until the temperature of the surface of its bulb is nearly as low as that of the dew-point ; that is, falls to that point at which air at the indicated temperature is so saturated that a farther lowering would be followed by condensation of moisture. As a matter of fact, the wet thermometer does not fall so far in a quiet air, since its bulb becomes surrounded by a layer of stagnant saturated air, and receives more or less heat from the surrounding warmer atmosphere. Again, in a saturated atmosphere, the wet thermometer may stand slightly higher than the dry one, owing to the fact that its covering protects it from loss of heat by radiation. GLAISHER'S TABLE. Reading of dry bulb thermometer. Factor. Reading of dry biUb thermometer. Factor. 1 Reading of 1 dry bulb 1 thermometer. Factor. 10 8.78 41 2.26 ' 71 1.76 11 8.78 42 2.23 ! 72 1.75 12 8.78 43 2.20 73 1.74 13 8.77 44 2.18 74 1.73 14 8.76 45 2.16 75 1.72 15 8.75 46 2.14 76 1.71 16 8 70 47 2.12 77 1.70 17 8.62 48 2.10 78 1.69 18 8.50 49 2.08 79 1.69 19 8.34 50 2.06 ; 80 1.68 20 8.14 51 2.04 81 1.68. 21 7.88 52 2.02 ' 82 1.67 22 7.60 53 2.00 83 1.67 23 7.28 54 1.98 , ; ) 84 1.66 24 6.92 55 1.96 ' ^- 85 1.65 25 6.53 56 1.94 86 1.65 26 6.08 57 1.92 87 1.64 27 5.61 58 1.90 88 1.64 28 5.12 59 1.89 89 1.63 29 4.63 60 1.88 90 1.63 30 4.15 61 1.87 91 1.62 31 3.60 62 1.86 ; 92 1.62 32 3.32 63 1.85 ! 93 1.61 33 3.01 64 1.83 ; 94 1.60 34 2.77 65 1.82 95 1.60 35 2.60 66 1.81 96 1.59 36 2.50 67 1.80 1 1 97 1.59 37 2.42 68 1.79 98 1.58 38 2.36 69 1.78 99 1.58 39 2.32 70 1.77 100 1.57 40 2.29 EXAMIXATIOy OF AIR. 255 For the purpose for which it is intended, the instrument is exposed until the wet thermometer ceases to fall, and then the reading of both is noted. From these data, with the assistance of Glaisher's factors (see table on page 254), the dew-point is easily calculated in the follow- ing manner : Multiply the difference in the two readings by the factor opposite the figure in the table corresponding to the temperature of the diy bulb, and subtract the product from this temperature. T.AJBLE OF TENSIONS. Tempera- ture. Corre- sponding Tension in inches of Tension Tempera- ture. Corre- sponding Tension in inches of Tension Fahrenheit. degrees C. mercury. in mm. Fahrenheit. degrees C. mercury. 1° —17.2° 0.046 1.17 51° 10.6° 0.374 9.50 2 —16.7 0.048 1.22 52 11.1 0.388 9.86 3 —16.1 0.05 1.27 53 11.7 0.403 10.24 4 —15.6 0.052 1.32 54 12.2 0.418 10.62 5 —15.0 0.054 1.37 55 12^8 0.433 11.00 6 —14.4 0.057 1.45 56 13.3 0.449 11.40 7 —13.9 0.060 1.52 57 13.9 0.465 11.81 8 —13.3 0.062 1.57 58 14.4 0.482 12.24 9 —12.8 0.065 1.65 59 15.0 0.500 12.70 10 —12.2 0.068 1.73 60 15.6 0.518 13.16 11 —11.7 0.071 1.80 61 16.1 0.537 13.64 12 —11.1 0.074 1.88 62 16.7 0.556 14.12 13 —10.6 0.078 1.98 63 17.2 0.576 14.63 14 ' —10.0 0.082 2.08 64 17.8 0.596 15.14 15 — 9.4 0.086 2.18 65 18.3 0.617 15.67 16 — 8.9 0.090 2.28 66 18.9 0.639 16.23 17 — 8.3 0.094 2.38 67 19.4 0.661 16.79 18 — 7.8 0.098 2.49 68 20.0 0.684 17.37 19 — 7.2 0.103 2.62 69 20.6 0.708 17.98 20 — 6.7 0.108 2.74 70 21.1 0.733 18.62 21 — 6.1 0.113 2.87 71 21.7 0.759 19.28 22 — 5.6 0.118 3.00 72 22.2 0.785 19.94 23 — 5.0 0.123 3.12 73 22!8 0.812 20.62 24 — 4.4 0.129 3.28 ; 74 2.3.3 0.840 21.34 25 — 3.9 0.135 3.43 75 23.9 0.868 22.05 26 — 3.3 0.141 3.58 76 24.4 0.897 22.78 27 — 2.8 0.147 3.73 77 25.0 0.927 23.55 28 2.2 0.153 3.89 78 25.6 0.958 24.33 29 — IJ 0.160 4.06 ! 79 26.1 0.990 25.15 30 — 1.1 0.167 4.24 80 26.7 1.023 25.98 31 — 0.6 0.174 4.41 81 27.2 1.057 26.85 32 — 0.0 0.181 4.60 82 27.8 1.092 27.74 33 + 0.6 0.188 4.78 ' 83 28.3 1.128 28.65 34 1.1 0.196 4.98 84 28.9 1.165 29.59 35 1.7 0.204 5.18 85 29.4 1.203 30.55 36 2.2 0.212 5.38 86 30.0 1.242 31.55 37 2.8 0.220 5.58 87 30.6 1.282 33.56 38 3.3 0.229 5.82 88 31.1 1.323 33.60 39 3.9 0.238 6.04 89 31.7 1.366 34.69 40 4.4 0.247 6.27 90 .32.2 1.410 35.81 41 5.0 0.257 6.53 91 32.8 1.455 36.95 42 5.6 0.267 6.78 92 33.3 1.501 38.12 43 6.1 0.277 7.04 93 33.9 1.548 39.31 44 6.7 0.28S 7.32 94 34.4 1.596 40.53 45 7.2 0.299 7.59 95 35.0 1.646 41.80 46 7.8 0.311 7.90 96 35.6 1.697 43.09 47 8.3 0.323 8.20 97 36.1 1.749 44.42 48 8.9 0.335 8.51 98 36.7 1.802 45.77 49 9.4 0.348 8.84 99 37.2 1.856 47.14 50 10.0 0.361 9.17 1 100 37.8 1.911 48.54 256 AIR Fig. 11. Having now determined the dew-point, the next step is to ascertain the elastic tension of the vapor present in the air, that is, the tension of the dew-point, and the tension of that necessary for saturation at the temperature of the dry bulb, which data can be obtained bv reference to the table on page 255. From these several data the relative humidity is calculated as fol- lows : Divide the tension of the dew-point by that of saturation at the actual tenn)erature, and multiply by 100. Example : Reading of drv bulb = 67° Reading of wet bulb =^ 62° Dew point = 67— (5X1-80) = 67—9 = 58° Relative humidity = ^' — ^ X 100 = 72.92 per cent. •' 0.661 ^ More accurate determination may be made by employing the "whirled" or "sling" thermometers. These are fastened to a string of such a length that the distance from the bulbs to the held end is exactly a meter. In use, they are whirled in a horizontal plane 100 times at the rate of one revolution per second. By their use, the errors mentioned as likely to occur when the observations are made in still air are elimi- nated. For all practical purposes, the use of the thermometers in the ordinary way gives sufficiently accurate results. In making determinations out of doors when the temperature is below the freezing- point, the wick may be dispensed with, and the bulb is then wetted by dipping it into Avater, the excess being removed by means of filter-paper or common blotting-paper, or water may be ap]>lied with a camel's-hair pencil. Below the freezing-point, however, the relative humidity is of little hygienic interest, sin^e^^he amount of moisture which air then can contain is but slight. A very convenient instruuient for quick approximate determinations without the necessity of tables and com]nitation is known as the hygrojihant of AVinlock and Iluddleston. It consists of a pair of ther- mometers and a cylinder, upon which is inscribed a series of 22 columns of figures numbered from 1 to 22, any one of which mav, by a turn of a knob, be brought into apposition with a fixed scale on the casing. (See Fig. 11.) To ascer- tain the relative humidity, note the difference in the readings of the thermometers, turn the cylinder, until the column liaving at its top the number corresponding to the diU'erence api)ears (t{»posite the scale, Hygrophajit. EXAMINATION OF AIR. 257 and read the figures opposite the number corresponding to the tem- perature of the wet bulb. Example : Eeading of dry bulb = 72= Beading of wet bulb ^ 60^^ Difference = 12"^ Fig. 12. The cylinder is turned until column 12 appears. Opposite 60 of the scale, the reading is 46 ; and this is approximately the percentage of saturation present. Determination of Carbon Dioxide-. — For the collection of samples of air for this determination, it is well to provide a number of bottles of about a gallon capacity. These should, first of all, be measured very carefully. This may be done by filling them with ice water and noting the number of 6c. required, or by determining by means of plat- form scales sensitive to 5 grams the difference between their weights empty and filled. It is well to place a distinguishing niunber and the figures denoting its capacity on each bottle, either on a label, or, better, by means of a writing diamond. AVhen used, the bottle shoidd be perfectly clean and dry. When it is necessary to employ the same bottle again, time being an object, the drying process is hastened very much by washmg first with water, then with a little alcohol to remove the small amount of water which will not drain away, and, finally, with a little ether for the re- moval of the residuum of alcohol. The small amount of adherent ether may then be removed by blowing a current of air into the bottle by means of a bellows. A number of tightly fitting rubber caps should be provided in place of corks or rubber stoppers, though if these are not at hand, the latter may be used ; but note should be made of the volume of air which they displace when they are inserted. Solutions RecLuired. — 1. Solution of Barium Hydrate. — Dissolve about 4.5 grams of barium hydrate and 0.5 of barium chloride in a liter of distilled water which previously has been boiled, in order to expel any carbon dioxide which it may contain. It is well to prepare an amount suf- ficient for future needs, say 4 liters, and to keep it in a bottle such as is shown in Fig. 12. This is provided with a rubber stopper with two per- forations, through one of which a bent tube, reach- ing to the bottom, and intended for withdrawal of the reagent, is inserted. Through the other is carried a tube extending only into the neck, and communicating at its outer extremity with a U-tube filled with pieces of pumice soaked while hot in a strong solution of caustic potash. The delivery tube carries at its outer 17 Bottle for barium hydrate. 258 AIR. end a piece of closely fitting rubber tubing, which is kept closed by means of a piuchcock. In withdrawing the reagent for use, a 100 cc. pipette is inserted into the free end of the rubber tube, suction is applied, and the pinchcock is opened. When the pipette is filled to the mark, the pressure is removed from the pinchcock and the pi]iette released. As the reagent is withdra\\'n, air flows in through the other opening, and is robbed of its carbon dioxide by contact with the caustic potash with which the pumice has been charged. This reagent is used for the absorption of the carbon dioxide contained in the sample of air under examination. The reaction is expressed by the following formula : BaO.Hj -r CO,=BaC03+H20. The function of the barium chloride is explained below. 2. Standard Solution of Oxalic Acid. — Dissolve 2.808 grams of pure oxalic acid in a liter of distilled water. One cc. of this solu- tion is equivalent to 0.5 cc. of carbon dioxide ; that is to say, will neutralize the same amount of barium hydrate as will combine with carbon dioxide to form barium carbonate. 3. Solution of Phenolphthalein. — Dissolve 0,5 gram of pheuolphthalein in 100 cc. of alcohol. This solution is used as an " indicator " of alkalinity. Process of Analysis. — The process of analysis depends u])on the fact that when a volume of the barium hydrate solutiou is brought into contact with carbon dioxide, its alkalinity is diminished by the forma- tion of barium carbonate, Avhich is a neutral body. The greater the amount of carbon dioxide to which it is exposed, the greater will be the reduction of its alkaline strength. A preliminary determination of the amount of oxalic acid solution which 100 cc. of the reagent Mill neutralize is made by titrating 25 cc. contained in an Erlenmeyer flask and colored by means of a few drops of the phenolphthalein solution, and multiplying the result by 4. After the reagent has been sub- jected to the influence of the gas in the air sample, a similar determin- ation is made. Tiie difference befvveen the two results, divided l)y 2, indicates the number of cc. of carbon dioxide present in the amount of air employed. The sample of air is obtained in the following manner : One of the bottles above mentioned is placed in the situation from which the air is to lie obtained, and its air content is displaced by means of a bellows provided at its outlet with a ruVjber tube of sufficient length to reach nearly or quite to the bottom. A half minute's jnunping is sufficient to insure that the original air is replaced by that under observation. One is sometimes admonished to be careful not to breathe in the direc- tion of the mouth of tlie l)ottle, but tliis is an unnecessary jirecaution, since the current issuing from the bottle is nnicli too jiowerful to admit of the entrance of any air except that propelled by the bellows. A much more and very necessary precaution to be observed is that the operator shall not allow his breath to reach the inlet holes of the bel- EXAMINATION OF AIR. 259 lows. After a half minute's pumping, the rubber cap is aftixed, and the bottle may then be carried to the laboratory, or, better, the treat- ment of the contained air may be proceeded with on the spot. An- other method of collecting the sample is often recommended in place of the one described. It consists in filling the bottle with water and emptying it where the an* is to be taken. By this process, the space originally occupied by water is filled with air, but the method is objec- tionable in that the water cannot drain away completely, and that that which remains serves to dilute, slightly it is true, the charge of barium hydrate next to be introduced, and thus brings in an error at the very outset. Xext, 100 cc. of the barium hydrate solution are introduced by drawing aside the edge of the rubber cap and inserting, into the opening so made, the point of the filled pipette, and allowing its contents to flow unaided into the bottle. The beginner will often incline inadver- tently to gain time, and assist the emptying of the pipette, by blowing into it, thereby vitiating his results with the impurities of his own respiration. As soon as the pipette is emptied, it is withdrawn and the edge of the cap is replaced. The bottle is then shaken thoroughly for about ten minutes, care being observed not to wet the cap, since in that event some of the reagent may escape by capillary attraction. At the end of that time it may be assumed that all of the contained carbon dioxid? has been brought into contact with and absorbed by the barium hydrate, which is then to be poured quickly from the bottle through a fairly large funnel into a glass-stoppered bottle of rather more than 100 cc. capacity. The solution, which is now more or less turbid from the presence of barium carbonate, is allowed to stand until, through settling of this substance, the supernatant liquid is clear. Three successive portions of 25 cc. each are next to be withdrawn by means of a pipette of the proper size, and, after addition of the indicator, titrated in Erlenmeyer flasks with the standard oxalic acid solution until the pink color caused by the former is made to disappear. So long as any color remains, one knows that barium still exists iii the form of hydrate, and that the contents of the flask are still alkaline, for phenolphthalein gives a pink tinge only in the presence of the alkalies. When the pink color disappears, the process is finished, and the reading of the burette is noted. The three portions of 25 cc. each are titrated in turn, and the mean of the results is multiplied by 4. The difference between this product and the figure obtained in the pre- liminary test of the strength of the reagent, divided by 2, indicates the number of cc. of carbon dioxide in the volume of air taken for analysis. In filling the 25 cc. pipette from the bottle containing the used reagent, great care should be observed not to stir up the sediment of barium carbonate. To perform the operation properly, it is necessary to insert the point of the pipette well below the surface, and to fill it up to the mark, or just beyond it, by one uninterrupted act of suction. If one stops to regain breath, part of the liquid already within the 260 AIE. pipette will escape downward during the interval with sufficient force to stir up the sediment. When the pipette is filled, the point of the tongue should be applied to its upper end, and the tip should then be withdrawn from the bottle. Then by placing the end of the forefinger over the opening of the tip of the pipette, the escape of its contents is prevented, while the forefinger of the other hand is replacing the point of the tongue. The reason for such careful avoidance of stirring up the sediment is that the presence of barium carbonate introduces a slight error in the titration. The slight excess of oxalic acid present when the color of the phenolphthalein is discharged attacks the sus- pended barium carbonate, forming barium oxalate and setting free the combined carbon dioxide. Thus : Hfi^O^ -^ BaCOj = BaCjO^ -f- H^O — CO^. The free carbon dioxide then attacks more of the carljonate and forms barium bicarbonate, which, being soluble and of alkaline reaction, causes the pink color to reappear. /OH co/ >Ba =BaH2(C03)j. co/ ^OH The reason for adding barium chloride in making the barium hvdrate solution is that most barium hydrate contains, in addition to small amounts of carbonate, traces either of caustic soda or of caustic potash. When either of these substances is brought into contact with barium chloride, mutual decomposition occurs, and we have as results barium hydrate and sodimn (or potassium) chloride. If the impurity were disregarded, it would cause errors, as shown Ijelow. The barium hvdrate solution when titrated Avith oxalic acid would behave according to the following formula : BaO^Hj + BaCOj ^ 2NaOH + 2H,C,0, = BaCA + ^aCOa -f Xa^CA - "IHjO (Barium (Barium (Sodium "^^ — fOxalic (Barium (Sodium hydrate) carbonate) hydrate) acid) oxalate) oxalate) In practice a very slight excess of oxalic acid is also present, and the reaction then proceeds still farther. The sodium oxalate attacks the bariimi carbonate, forming barium oxalate and sodium carbonate. Thus Na-^CjO, -L BaCOg = BaCA + Na^COj. Next, the sodium carbonate neutralizes the traces of free oxalic acid, and any surplus causes a reappearance of the pink color and neces- sitates farther addition of oxalic acid. This causes the formation of more sodium oxalate, which in its turn attacks another portion of the barium carbonate, with the same results as before ; and so the cycle continues until the last trace of suspended carbonate is decomposed. If the hydrate contains no impurities, the addition of chloride is unnecessarv. EXAMINATION OF AIR. 261 Corrections. — In figuring the results of the determination, certain corrections are necessary. First, the volume of the barium hydrate used (100 cc.) must be subtracted from the capacity of the bottle, since its introduction displaces an equal volume of air ; and next, allowances must be made for any departure from standard temperature and baro- metric pressure, since the capacity of the sample bottle is reckoned for air at 0° C. and 760 mm. pressure. In order to make the necessary corrections for temperature and pressure, tbe thermometer and barom- eter should be noted at the time of taking the sample. In determining the amount of correction, we are guided by two physical laws : that for each degree of temperature, air expands a con- stant fraction of its own volume (La^v of Charles) ; and that the vol- ume of a gas is inversely proportionate to the pressure (Law of Boyle). For each degree centigrade above or below 0° C, air expands or con- tracts 0.0036648 of its volume ; and this figure is known as the co- efficient of expansion for centigrade degrees. For each degree Fahren- heit above or below 32°, air expands or contracts 0.002036 of its volume ; and this is known as the coefficient of expansion for Fahren- heit degrees. Thus, 1 liter of air, heated to 40° C, will expand to 1 + (40 X 0.0036648), which equals 1.146592 liters; or heated to 104° F. (104° F. = 40° C), it will expand to 1 + (72 X 0.002036), which equals 1.146492, as before. Again, the same volume cooled to — 15° C, will contract to 1 — (15 X 0.0036648), or 945 cc. ; or cooled to 5° F. (5° F. = 15°C.), it will become 1 —(27 X 0.002036) or 945 CO., as before. So an apparent volume of 1,000 cc. at any temperature above freezing is in reality a smaller volume expanded to that size ; and at any temperature below, is a larger volume brought to that size by contraction. To correct volume for temperature, we must divide the apparent volume by 1 plus the product of 0.0036648 times the number of degrees away from 0° C, or in case of temperatures below freezing, by 1 minus that amount. If the Fahrenheit scale is used, the appro- priate coefficient and factors must be substituted. Thus we may employ a set of formulge as follows : For temperatures above 0° C V — For temperatures below 0° C V- For temperatures above 32° F. . . . V - 1 + 0.0036648.^° C. V 1 — 0.0036648i° C. . F^ ■ 1 + 0.002036 (<° F. — 32) For temperatures below 32° F. . . . V- 1 — 0.002036 (32 — <° F.) In the above, V = correct volume. V ■= apparent volume. Inasmuch as volume is inversely proportionate to pressure, the true volume at any observed pressure is obtained by multiplying the apparent volume by the barometric pressure expressed in millimeters 262 AIR. or inches, and dividing the product by 760 or by 29.92, as the case may be. We may use, then, this formula : 760 Applying it, we find that an apparent volume of 1,000 cc. at 750 mm. becomes l,000X7_50_9g.^^. 760 or using the other scale, the barometer standing at 29.53 inches (29.53 in., 750 mm.), it becomes 1,000X29^ _ 987 ec. 29.92 If the barometer reads higher than the standard pressure, the true volume will be greater than the apparent. Thus, suppose the pressure to be 30.22 inches, then 1,000 cc. will represent 1,000X30.22 1^1,. —^ — =^ 1,010 cc. 29.92 ' Instead of going through two separate calculations, we may make both corrections at once by means of one formula which is a combina- tion of the two kinds already used. For temperatures above 0° C. the correct volume is obtained by means of the following : V= Y'XB, (1 + 0.0036648.<°) 760 By changing the plus sign to minus, the formula is adaj)ted to tem- peratures below freezing. If the Fahrenheit thermometer is used, and the barometric pressure is expressed in inches, the formula is as fol- lows : V= E'_X£ , [1 + 0.002036 {t°F. - 32)] 29.92 and if the temperature is below 32°, it must be changed to V=- y X B [1 — 0.002036 (32 — <° F.)] 29.92 In these formulae : y == correct volume. V =r a])parent volume. J5 = barometric pressure. t° = tcm])('rature. In order to avoid the tedious process of multiplication and division which the working of these formula involves, recourse may be had to the admirable tables of Dr. AYalter Hesse,' wlierein can be found the correction to be made for all teni])('raturos between — 2° and ."iO^ C. and for all pressures between () the following lonnula : 2KI -f- H,,0 + O3 = 2KOII + Oj + I2. This reaction is the basis of most of the ])rocesses which have been proposed for qualitative aud (juantitative determination, none of which may be regarded as of value, since there are many sources of error to ' Comptes rendus, ('XX VI.. \>. T4(). •-' Ibidem, CXXVI,, pp. S71, 1)31, 97.3. s Thidein, p. 038. EXAMINATION OF AIR. 267 be taken into account, sources impossible to eliminate and of impor- tance impossible to compute. The presence of ozone in the air is supposed to be demonstrated when, on exposure of paper saturated with starch paste containing potassium iodide, a blue color gradually develops, owing to the action of the liberated iodine on the starch. Quantitative determinations are made by comparing the tint with a standard scale, the depth of color being dependent upon the amount of iodine liberated, and this upon the amount of ozone present. The papers are prepared in the follow- ing manner : From 2.5 to 10 grams of starch are taken, according to the recommendations followed, and, after trituration with a small amount of cold water, are boiled for about ten minutes in about 200 cc. of water, and filtered. One gram of potassium iodide iu solu- tion is next added gradually with constant stirring. Strips of stout filter-paper, wet with distilled water, are soaked in the starch prepa- ration until they are thoroughly impregnated (about two to four hours), then removed with the aid of forceps, spread flat, and dried in the dark. AVheu used, they are hung up out of the direct sun- light and exposed for a definite time, then removed, moistened with water, and compared with the scale. The objections to the process are that a number of other substances which may be in the air, such as certain volatile organic acids, chlorine, nitrous acid, and hydro- gen peroxide, cause this same chemical reaction ; that the blue color is destroyed by other substances, as sulphuretted hydrogen and sulphur- ous acid ; and that light, moisture, heat, and wind exert very decided modifying influences. Thus, wind brings more air into contact, sun- light bleaches the color, moisture hastens the bluing, and heat dissi- pates the free iodine. In order to diflPerentiate between ozone and nitrous acid, it has been proposed to use neutral litmus (violet) paper instead of ordi- nary filter-paper in making the strips. The KOH formed in the reaction will change the violet to blue, while nitrous acid, chlorine, and organic acids will convert it to red, or bleach it, or leave it unchanged. In spite of the fallacies mentioned, the weight of evidence thus far obtained iu ozonimetry shows that the reaction with starch is most marked in pure air at the seashore and at great heights, and that but little reaction occurs indoors. Determination of Dust. — Dust is determined quantitatively in two ways, and the results are expressed in terms of weight or of number. In order to ascertain the iceight of the dust contained in a given volume of air, a chloride of calcium tube, containing per- fectly dry absorbent cotton or glass wool, is weighed accurately, and then attached to a water suction-pump with an air-meter between. A large amount of air, say 500 liters, is then drawn through as quickly as possible. When a sufficient amount has passed, the tube is detached and placed either in a drying-oven or in a desiccator over sulphuric acid, and kept until it ceases to lose weight (moisture). The 268 AIR. net increase in weight represents the amount of dust in the volume of air aspirated. To determine the number of dust particles in a given volume, the method of Aitkin is employed. The apparatus includes a shallow metallic box with glass top and bottom etched in squares. Into this box, containing air which has been freed from dust by filtration through cotton, and is kept saturated with moisture by means of wet filter- paper, a small measured amount of the air under examination is in- troduced. By causing the formation of a partial vacuum, each par- ticle of dust becomes coated with condensed moisture and hence tends to fall upon the etched squares of the bottom. The number deposited is counted with the aid of a magnifying glass. The number of par- ticles varies, according to Aitkin's observations, from 8,000 to 100,000 per cubic inch in the country, and from 1,000,000 to 50,000,000 in cities. Bacteriological Examination. — The method which involves the least trouble and requires a minimum of apparatus, and which for all practical purposes gives greatest satisfaction, consists in exposing^ gelatin plates or Petri dishes for a definite period, and then covering them and letting the colonies develop. After the proper interval, the number of growths may be counted, and the individual species isolated and studied. This method is very useful for comparative work, the results being given as the number of colonies which develop after a given exposure. For more accurate quantitative work, Petri ' devised a process of sand filtration. A glass tube, 9 by 1.6 cm., serves to carry two small filters, whieli are arranged in the following manner : Two small tightly fitting dia[)]n'agms of fine wire gauze are inserted into the tube at a point midway between the ends. Into one side, a quantity of fine quartz sand is i)acked, and upon it, to keep it in place, another dia- phragm is driven. A\)oye this, the space is filled with a cotton plug. The tube is now reversecl and a second filter of sand is made in the same way. After complete sterilization, the cotton plug in one end gives way to a rubber stop])er with a single ]ierforation, through which passes a glass tube connected with an aspirating pump. The other cotton jilug is removed and the process of suction begun. When a sufficient amount has been drawn through, the two filters are removed, eacli by itself, and mixed with the nutrient gelatin from which plates are next to be made. The first filter should contain all of the organisms, the second serving as a control. Ficker suggested an improvement in the construction of the filters, substituting for sand, which to a certain extent, masks the colonies, powdered glass, which has not this disadvantage. A still better material is fine sugar, the use of which was suggested first by Sedgwick. The advantage of this is that it is dissolved in the liquefied gelatin, and thus disappears from view, and, therefore^ neither masks the colonies nor can be mistaken for them in counting, ' Zeitschiift fiir Hygiene, III., p. 1. EXAIIIXATIOy OF AIR. 269 Seclg^vick^s method of collecting organisms aucl obtaining cultui'es is -one which, on the whole, is preferable to any other that has been sug- gested. His apparatus, known as the aerobioscope, is a glass tube about 14 inches in length, shaped like a hydrometer and open at both ends. The narrow portion, which is rather less than half the length of the tube, has an internal diameter of 0.2 inch ; the broader portion has an internal diameter of 1.8 inches, and at its free end is constricted for an inch to about half its size. Into the outer end of the narrow por- tion, a diaphragm consisting of a roll of fine wire gauze is inserted to act as a plug for the sugar filter. The two open ends are stopped with cotton, and the apparatus is then sterilized. The plug at the larger end is next remoyed and the sugar, sufficient in amount to fill the small tube above its contained diaphragm, is introduced. The plug is replaced, and then the whole is sterilized at 120° C. for several hours. In use, the apparatus is held in a vertical position with the narrow portion down, the plugs are removed, and a measured volume of air is drawn through by means of an aspirating apparatus connected by a rubber tube to the lower end. When the desired amount of air has been aspirated, the -sugar with the bacteria which it has arrested is brought, by proper manipulation, into the broad part, into which, by means of a bent funnel, a sufficient amount of liquefied nutrient gelatin is introduced. The plug is replaced, and the tube is then roUed and chilled on ice, and set aside for the development of colonies. After the proper interval, the count is made in the usual manner. The methods above given have generally superseded that of Hesse, who was a pioneer in this branch of investigation. His apparatus con- sists of a glass tube, 28 inches long and about IJ wide, supported in a horizontal position upon a wooden tripod. One end is covered with two rubber caps, the inner of which has a single perforation; the other end is closed with a rubber stopper with an outlet tube of glass jjlugged at €ach end with cotton and connected with a pair of aspirating flasks of a liter capacity. The tube is sterilized and charged with 50 ce. of gelatin, which is allowed to solidify before use. In conductmg the operation, the outer cap is removed, thus exposing the inner perforated one, and a current of air is drawn slowly through by the action of the aspirating flask, which, filled with water, empties itself into the other. By reversing the flasks, any number of liters of ah^ may be drawn through. In its passage, the air deposits its bacteria on the gelatin. The process has many disadvantages, and can make no great claim to accuracy. CHAPTER III. THE SOIL. Notwithstanding the constant and necessarily intimate relation of all life to the soil upon which we build our habitations, from which we derive in such great part our supply of drinking-water, into which we cast vast quantities of organic filth, and to which we consign our dead, the subject of the sanitary importance of the soil has not until within comparatively recent years received the attention which it merits. That the soil exerts important influences on the public health, was recognized long before the time of Hippocrates^ and extensive researches on the subject figure among the earliest investigations of the modern hygienist, but by far the greatest part of the attention paid to the study of the soil has been due to con- siderations of ]iublic wealth rather than of public health. With the gradual development, however, of a more accurate knowledge of the causes of disease, has come an increasing interest in the rela- tions of the soil to those causes, and what has hitherto ))een a rather neglected Held of exploration now bids fair to be well and thoroughly tilled. That portion ol" the earth's ci'ust in which we as liygienists are interested includes the su])crficial layer, known as tilth or arable soil, which is the result of the disintegration of rocks and decay of animal and vegetable life, and the subsoil, which lies directly beneath. The former varies from a few inches to several feet in depth ; the latter extends few or many fWt downward to the hard})an or other imper- meable stratum. Soil is a mixture of sand, clay, and other mineral substances, with humus, or organic matter, and living organisms; and it is classified according as one or another of its constituents predominates. The usual classification of soils includes sands, clays, loams, marls, luunus, and peats. Sandy .soils consist almost wholly, or at least more than four-fifths, of pure sand of any kind. Clays are stiff soils consisting chiefly of silicate of aluminum and other very finely divided mineral matters. Clay exists in particles of the smallest possible size, is very cohesive, possesses a high degree of plasticity, and plays a very important ])art in determining the fertility of soils, their texture, and their capacity fi)r holding water. Its ])las- ticity is due to the })resence of a small ]>r()|)ortion of hydrated silicate, and is modified very greatly by the jidditiou of less than a hundredth part of caustic lime. It is exceedingly impermeable to water, and when wet dries with great slowness. 270 THE SOIL. 271 Loams are mixtures of sand, clay, and humus ; hence their proper- ties partake of the characteristics of these substances according to the extent to which each is present. When sand predominates, they are designated as light ; and when clay prevails, they are known as heavy. These terms, however, have no reference to weight, but to the ease or difficulty with which they are worked in the processes of agriculture ; and, indeed, those soils which are the lightest in this sense are the heaviest in actual weight. Since loams consist of varying proportions of the chief constituents, it is obvious that the word loam may have but little significance without some qualifying term, and they are, there- fore, divided into five classes, as follows : 1. Heavy clay loam, containing 10-25 per cent, of sand. 2. Clay loam, containing 25-40 " " 3. Loam, containing 40-60 " " 4. Sandy loam, containing 60-75 " " 5. Light sandy loam, containing 75-90 " " Mixtures containing less than 10 or more than 90 per cent, of sand are classed, respectively, as clay or sand. Marls are mixtures of clay, sand, and amorphous calcium carbonate in various proportions, and contain, often, potash or phosphates from the fauna and flora of the sea. From their content of carbonate of calcium they are known often as lime soils, and according as one or another constituent predominates they are designated as clay marl, sand marl, and shell marl. All contain varying amounts of humus. Humus is a term used to designate the entire product of vegetable decomposition in the various intermediate stages of the process. It is the essential element of vegetable mould, and is necessarilv of most complex composition — so complex, indeed, that it cannot definitely be determined. It is composed of a great number of closely related definite chemical compounds, chief among which are ulmin and ulmic acid, which are supposed to characterize brown humus ; humin, and humic acid, which dominate dark, or black humus ; and crenic and apocrenic acids. Its principal characteristic is its high percentage of nitrogen, especially marked in some of our prairie soils and in the " black soil " found in the provinces of the Ural Mountains, which, accorduig to Von Hensen,^ contains as much as from 5 to 12 per cent, of organic matter. Its complete decay is most rapid in warm well- drained soils permeable to air, and in such soils the amount of humus present at any one time will be relatively small, whUe m soils which are damp, not well ventilated, and, for months at a time, frozen, its accumulation is favored. While its ultimate products of decay are of the greatest importance to vegetable growth, it does not follow that its complete absence renders a soil necessarily sterile, or even poor, provided the necessary nitrogen is supplied in the form of nitrates. But its presence is necessary to the growth and life processes of the soil bacteria, without whose assistance many plants would fail to thrive. Peat, muck, and humus soils contain large amounts of humus, but differ ^ Zeitschrift fiir wissenschaftliche Zoologie, XXVIII., p. 360. 272 THE SOIL. according to the conditions under which they are formed. Peat and muck result from the incomplete decay of vegetable matter under water ; the former term applies to that which is compact and fibrous ; the latter is less compact, not fibrous, and, when dry, easily reduced to powder. They contain but a small amount of mineral matter. Humus soils are soils which contain large percentages of vegetable mould with ordinary soil constituents. The expression )-ocky soil applies to any kind of soil containing masses of rock. Gravelly soils are those which contain notable amounts of gravel, which consists of small fragments of rock more or less worn by tlie action of water, and larger and coarser than sand. Alkaline or salt soils are soils which contain considerable amounts of soluble salts, especially carbonate and sulphate of sodium and salts of calcium. Constituents of the Soil. — The chief constituent of the soil is silica, which, it is estimated, forms about two-thirds of the entire earth's crust. Next in abundance is aluminum, chiefly in the form Ox" clay (silicate of aluminum). Lime and magnt\'^ia are large constituents, existing cliiefly as carbonates in the form of" limestone. Both are indispensable to the growth of plants, and lime exerts a marked inflluence on the physical condition of the soil and upon the processes of nitrification. Although its principal com})inati(>n is carbonate, it exists also largely as phos- phate and sulphate. Iron is universally present, and is of veiy great importance to vege- tation, although but a small amount is needed. The red and yellow- colors of soils are due to the presence of iron compounds. Manganese stands second to iron in abundance among the heavy metals, but is of much less importance. It is a constituent of many ])lants, notably of tea. Chlorine is not^a large constituent ; it occurs chiefly in combina- tion with sodium, potassium, and magnesium. Its total amount in ordinary unjioUuted soil seldom exceeds yo^^-qit P'*^'^ ^^ *^^^ whole. Sul- phur occurs as sulphides and sulphates, the latter usually in combina- tion with calcium. It is very necessary to vegetal)lc growth, as it is an essential element of vegetable albumin. Phosphorus in the form of phosphates of lime, magnesia, iron, and alumina, is another essential element, widely distributed in small amounts. Sodium and potassium are present, chiefly in the form of insoluble silicates and partly as chlorides. Their total in combination seldom exceeds 4 per cent. Nitrogen exists in soils in three distinct forms : proteids, ammonia and its salts, and nitric acid and nitrates. In average soils, the total nitrogen is not large in amount — considerably less than 1 per cent. — but in some exceptionally rich humus soils 4, 5, and even 6 per cent, are found. In the organic combinations (jn'oteids) it is not available as plant food, consequently these must be broken up into simpler forms in order to be of direct use. In their decomposition, the second form, ammonia, is produced, but not all the ammonia of the soil is from this source, for some is broug-ht into it from the air bv rain. And in the THE SOIL. 273 second form, also, it appears to be not available as plant food, but even, according to Bouchardat and Cloez,^ seems to act as an energetic poison when absorbed bv plant roots from solutions of 0.1 to 0.01 per cent, strength. So it is probable that complete oxidation to the third form is necessary for the absorption of any form of nitrogen. As soon as the ammonia is oxidized in its turn to nitric acid, this latter combines with sodium, potassium, or calcium, and the resulting nitrates are then ready for absorption. All of these changes from the complex proteid to the simple nitrate are carried along by different groups of micro-organisms, but no great accumulation of the end products occurs, because, while vegetation is flourishing, they are removed as fast as formed, and when it has ceased, they are washed down into the subsoil by the rain and melting snow. The amount of organic matter in soils varies widelv according; to circumstances, but the amount necessary for vegetation is Cjuite small, although certain crops, as tobacco and wheat, requu'e much more than others, as oats and rye. The soils richest in organic matter are the peats and mucks ; next come the very rich humus soils, which may yield more than a fourth of their weight. From 10 to 15 per cent, denotes unusual richness, and about 6 per cent, may be regarded as a fair amount for a productive soil. Physical Properties of Soils. — Pore-volume. — In all soils, no mat- ter how closely the individual particles are packed, there must exist a greater or less amount of interstitial space, which may be filled with water or air, or both together. The sum total of these interstitial spaces is known as the jjorosity or pore-volume, and is expressed in per- centage of the volume of the soil. Its amount depends not upon the size of the soil particles, but upon their uniformity or lack of uniform- ity of size, and upon their arrangement. If we have, for instance, a Fig. 15. Fig. 16. very coarse soil, consisting of particles of uniform size as large as peas, and another of uniform particles the size of small shot, we shall find, on determining their pore-volume, that it is practically the same in each case, and is probably not far from a third of the whole. Packed in the most solid manner possible, which is that in which each sphere rests on three beneath it (arranged like the familiar pyramid of mar- bles), helps support three in the layer above it, and comes in contact Anth others at six equidistant points along its equator, as in Fig. 15, the volume of interstitial space will equal 25.95 per cent, of the whole. Packed as loosely as possible, so that each rests upon but one, sup- ^ Deutsche medicinische AVoehenschrift, 18S6. IS 274 THE SOIL. ports another, and comes in contact with but four of its neighbors in the same layer as itself, as in Fig. 16, the volume of the interstices will be 47.64 per cent.' Thus a soil composed of spherical grains of uniform size would have, regardless of the coarseness of the grains, a pore-volume of not less than 25.95 per cent. That the size of the individual grains makes no difference, may easily be demonstrated in a practical manner. If we take two cylindrical glass vessels of the same size, fill them to the same height with water, and then add to the one a measure of large shot and to the other an equal measure of much finer shot, and secure as solid packing as pos- sible by gentle tapping, it will be found that the Mater in each cylin- der has risen to practically the same height ; that is, that the actual volume of each is about the same. There will be, perhaps, some slight difference one way or the other, owing to the impossibility of securing absolute uniformity of packing, and to the error due to the inequality of the spaces along the circumference of the cylinders. But in nature we do not deal with perfect spheres or with soils made up of particles all of the same size, but with soils composed of angular pieces of varying size. The greater the variation in size of the particles, the greater the possibility of variation from the limits of pore-volume as given above, ^yith varying size, the small particles may fall into the spaces made by the larger ones, and the spaces between the ne^' comers may be trespassed upon by still smaller grains, and so on until the interstitial space has been reduced to a minimum. To illustrate this diminution in a practical way, fill a large beaker with marbles, then pour into it, from a graduate, sufficient water to displace all of the air in the interstices, and note the amount of Mater required, Mhich is the pore-volume of the mass. Next, pour out the water as completely /^s possible and run on to the surface of the marbles a quantity ofvcoarse sand or shot, and shake the vessel gently in all directions so as to favor their descent into the spaces beloM'. When all have penetrated that can, pour in Mater ag-ain until it ap- pears at the surface, and note the amount required ; this is smaller than before, on account of diminished air spaces. Now pour off the Mater a second time, add still finer shot, and repeat the operation as before. So long as new matter can be added, so long Mill the pore- volume shoM' a diminution. Irregularity of size and shape of the particles may also have an in- fluence in the other direction, and cause the formation of large spaces and increased pore-volume. All soils, even the most compact rocks, have a certain amount of pore-volume, and some apparently compact masses, such as sandstone, have as much as 30 per cent. In soils which are cemented into homo- geneous masses, the pore-volume sinks to a minimum, but in ordinary soils it amounts to about 40 per cent. Permeability of Soils. — The permeability of a soil to air depends not, as it might appear, upon the amount of its pore-volume, but ujion ' Soyka, Der Boden, Leipzig, 1887. THE SOIL. 275 the size of the individual spaces. In fact, a soil of high pore-volume may be almost impermeable to air in comparison with one of less pore- volume, as will be shown ; and the pore-volume is of itself no measure whatever of permeability, Avhich diminishes in an extraordinary degree with diminution in the size of the soil particles. The greater the num- ber of the individual spaces, the greater the number of angles and the greater the friction of the entering air ; and, conversely, the less the number, and consequently the larger the size of the spaces, the less the number of angles and the less the obstruction. A series of experiments conducted very carefully by Renk ^ with different kinds of soil in cylinders of equal height, through which air was forced under the same degree of pressure, yielded the following interesting results : Nature of soil. Diameter of grains. Pore- voluaie 5i. Pressure in mm. of water. Amount of air in liters per minute. Ratio. Fine sand Medium sand . . - Coarse sand .... Fine gravel .... Medium gravel . . . Less than ^ mm. J to 1 mm. 1 to 2 mm. 2 to 4 ram. 4 to 7 mm. 55.5 55.5 37.9 37.9 37.9 20 20 20 20 20 0.00133 0.112 1.280 6.910 15.540 1 84 961 5,195 11,684 Thus it is seen that a fine sand with a pore- volume of 55.5 per cent. permitted the passage of but 1 volume of air, while a gravel of medium coarseness with much lower porosity permitted the passage of 11,684 times as much in the same uuit of time. Renk showed, farther, that with soils of the finer textures, permeability to air is directly pro- portionate to pressure, but that this is not true of those of coarser grain. Height Units Ratio of Kature of soil. Size of grain. of of volume of air column. pressure. passed. Fine sand Less than J mm. 0.50 m. 1 1.5 1 1.5 Medium sand J to 1 mm. 0.50 m. 1 2 3 1 2 3 2.00 m. 1 2 .3.6 1 2 3.6 Coarse sand 1 to 2 mm. 0.50 m. 1 2 3 1 1.91 2.78 2.00 m. 1 2 3 1 2 2.9 Fine gravel 2 to 4 mm. 0.50 m. 1 2 4 1 1.67 2.30 2.00 m. 1 2 3 1 1.77 2.42 Medium gravel 4 to 7 mm. 2.00 m. 1 2 3 1 1.65 2.19 ^ Zeitschrift fiir Biologie, XV., p. 205. 276 THE SOIL. The abseuce of any connection between pore-volume and permeabil- ity has been shown also by von Welitschkowsky/ from whose results the following: table has been constructed : Kature of soil. Pore-volume ^. Pressure in mm. of water. Amount of air in liters Ratio, per minute. 41.87 40.64 37.38 35.47 50 60 50 50 0058 1 Medium sand C'oarse sand Fine gravel 0.8990 155 7.399 l,'i76 33.651 2 5,802 1 Since permeability diminishes with fineness of texture, it follows that clay and similar soils possess this proj^erty in the smallest degree, and that when these are mixed with sandy soils they must necessarily lessen it to a very marked extent. But clays and loams may occur in very open crumbly form, that is, in loose fragments of varying size, each consisting of myriads of small particles held together by the aid of moisture ; and such soils show a high permeability, due to their large interstitial spaces. The degree of })ermeability to air is influenced very greatly by the amount of contained moisture, the maximum influence being exerted by decided wetness. This is due to the fact that the greater the amount of water present in the mterstices, the greater the diminution in the space available for the passage of air and the greater the obstruction to its movement. Thus the complete occlusion of the interstices by water is equivalent to absolute impermeability, except when the pressure of air is sufficient to displace the water and move it along. V In the case of soils that are only j)artially wet, the diminution in permeability varies according as the moisture enters from above by rain or from below by capillary attraction from the water in the sub- soil. This is ownng to the fact that when the soil is -wetted from above by rain, the su])erficial interstices are occluded movo or less completely, and the air in those below is restrained in its movement ; while when the moisture is derived by capillary attraction, the air is displaced upward, and the superftcial interstices are more or less com- pletely open. The action of downward and upward moistening has been investigated by Renk,'^ whose results, in part, are given in the following table : ' Beitrag zur Kenntniss dor PermeaV)ilitat des Bodens fiir Luft : Archiv fiir llvgiene, II., p. 483. '•'The height of the cohiiim of material in this experiment was three-fourths of a meter, instead of a half, as in the ease of the three others. "With an ecjual height the result would have been much larger. ^ Loco citato. THE SOIL. 277 Nature of soil. Pore- Yolume 'ft. Moisture. Pressure. Eatio of air passed. Medium gravel Fine gravel . Coarse sand Medium sand . Fine sand 37.9 37.9 37.9 41.5 55.5 55.5 absent from above from below absent from above from below absent from above from below absent from above from below absent from above from below absent from above from below 20 20 20 40 40 40 40 40 40 150 150 150 150 150 150 150 150 150 15.54 14.63 13.70 14.04 13.16 12.55 2.33 1.91 1.71 0.57 0.11 0.00 0.04 0.03 0.00 0.01 0.00 0.00 Permeability is lessened also by freezing temperatures, by reason of the fact that the contained moisture expands about one-eleventh of its volume as it freezes, and so occupies that much more space in the inter- stices. Moreover, when frozen, the moisture is in a fixed rather than a movable condition, and causes the production of a compact mass more or less resembling stone. The finer the grain, the more solid the product, and the greater the diminution of permeability. Renk ^ deter- mined the diminution in the permeability of soils of different grain size due to freezing, as follows : Nature of soil. Source of moisture. Permeability. Diminution. Moist. Frozen. Medium gravel a u Fine gravel Coarse sand from above " below " above " below " above " below " above " below '' above " below 14.63 13.70 13.16 12.55 1.91 1.71 0.11 0.00 0.23 0.00 13.87 12.20 12.54 10.18 1.64 1.27 0.07 0.00 0.00 0.00 5.2/^ 10.9 5.4 19.0 14.1 25 7 Medium sand (< a 36.4 11 a 100 u u The degree of permeability of soil to water, like that of permeability to air, is governed by the texture rather than by pore-volume, as is shown by the following results obtained by von Welitschkowsky,^ who determined the rates at which water would pass through columns of soil of differing fineness packed in cylinders of equal diameter. Each ' Beitrag zur Kenntniss der Permeabilitat des Bodens fiir Luft : Archiv fur Hygiene, II., p. 483. ^ Experimentelle Untersucbung iiber die Permeabilitat des Bodens fiir Wasser, Archiv fiir Hygiene, II., p. 499. 278 THE SOIL. specimen was first completely saturated and then kept so during each experiment, the water supplied being kept at constant level. Height of column of soil. Height of water column above soil surface. 20 cm. 50 cm. 20 cm. 1 50 cm. Nature of soil and size of grain. Amount of water discharged in liters per min. Fine sand, less than J mm Medium sand, J to 1 mm 0.00024 0.175 1.767 8.570 14.909 0.00059 0.435 4.014 16.190 0.00014 0.123 1.351 7.465 12.872 0.00028 0.237 2.422 Fine gravel, 2 to 4 mm Medium gravel, 4 to 7 mm 11.705 Comparing these results with those obtained by the same investi- gator in his experiments on permeability to air, it will be noticed tliat the total pore-volume has here even less significance. Nature of soil. Fine sand . . Medium sand Coarse sand . Height. Pressure. Pore- volume ^. 50 cm. 50 cm. 50 cm. 50 cm. 50 cm. 50 cm. 41.87 40.64 37.38 Katio of permeability. To air. To water. 1 155 1,276 8,650 Capacity for Water, and Water-retaining Capacity. If to a vokime of any soil packed into a cylinder of gla.-^s or metal we add water in such a way that all of the air in the interstices is dis- placed, the soil is then saturated and the amount of the contained water represents the total " water capacity," which, it is seen, equals the pore-volume. The "water-retaining capacity," is quite another thing, and depends upon the structure and composition of the soil, and, in a minor degree, upon other considerations. If for the imper- vious bottom we substitute one of wire gauze or coarse cloth, the con- tained water will begin to drain away, owing to the force of gravity, and the flow will by degrees become less and less, and finally cease. Then the interstices, which originally were filled with air alone and next with water, are filled in part with the one and in part with the other. Bv com ))a ring the original weight of the volume of soil with its weight in its now wet condition, its power to retain water is easily de- termined. This power is the result of two forces acting in opposition to the force of gravity ; namely, surface attraction of solids for liquids, and capillary attraction. The water which is simj^ly adherent to the surfaces of the soil grains is known as hygroscopic water, while that which is held in the capillary si)aees is called capillary water ; and it is the latter which, in any but the coarsest soils, constitutes by far the larger part of the retained moisture. Not all of the interstices of a soil form capillary spaces, but only those of which the Iwundary walls are separated only by intervals CAPACITY FOR WATER, AND WATER-RETAINING CAPACITY. 279 which come within the limits of capillary magnitude. Thus, a coarse soil may contain comparativelv few such spaces, while one of a fine texture may have its particles so closely approximated that all of its interstitial spaces are capillary. It follows, therefore, that compact soils possess greater retaining power than those with large inter- stices which permit rapid percolation, and that when the texture is so fine that all the spaces are capillary, the maximum retaining power is attained. The influence of soil texture on capacity for holding water may be seen in the following table of some of the results obtained by Hof- Diameter of grain in mm. Pore-volume per 1,000 CO. Amount of con- tained water in ee. Amount of con- tained air in cc. Per cent, of pore- volume filled with water. 5 434 55 379 12.7 3 41.8 77 341 18.4 2 410 98 312 23.9 1 400 150 250 37.5 0.5 413 270 143 65.4 less than 0.5 413 347 66 84.0 The water-retaining capacity of soils is determined very largely also by the amounts of organic matter present ; a soil rich in organic matter will, other conditions being the same, show more water than another of less richness. The extreme influence is observed in the case of humus, which can hold ten times its weight of water. In view of this influence of organic matter, it is very clear that one way to help keep a soil dry is to avoid discharging filth into it, and thus keep it clean. For the purpose of illustrating the influence of very fine soil par- ticles (clay) and of organic matter (humus), the following results of an investigation by Wolif ■ may be quoted. He packed soils of varying clay and humus content into a metallic vessel with a permeable bottom, saturated them completely with water, then superimposed a column of water of equal cross-section and 8 cm. high, and observed the time required for the added water to be delivered below : Percentage of clay. Percentage Time required. Nature of soil. of Per cent. Ratio. humus. Hours. Ratio. Verv fine sandv loam .... 15.74 1. 0.88 20.3 1. Verv fine sandv loam .... 15.96 1- + 1.40 25.8 1.27 Black, rich, chalkv loam 18.17 1.15 6.87 31.0 1.52 Very fine sandv loam .... 25.93 1.64 0.92 75.8 3.73 Very clavev soil 42.56 2.70 0.66 133.0 . 6.55 Soil with considerable clav . . 29.76 1.89 2.19 188.0 9.26 It will be observed that the soil which permitted the passage of the water in the shortest time was poorest in clay and almost so in humus, ^ Archiv fiir Hygiene, I., p. 273. ^ Anleitung zur chemischen Untersuchimg landwiithschaftlich wichtiger Stofte, 1875, 280 THE SOIL. and that the one which required the longest time combined a consid- erable amount of clay, not the highest, with a high percentage of humus, also not the highest. The highest percentage of clay was associated with the lowest amount of humus, and the highest of humus with a low content of clay ; but these two soils (Xos. 5 and 3) were both less impermeable than that (No. 6) which contained less clay than the one and less humus than the other. It is to be noted, however, in the case of the soil with the highest proportion of clay and lowest of humus, that it contained 12.8 per cent, of chalk as against 2.28 per cent, in the most impermeable, and that this substance, as has been mentioned, has a very great influence in diminishing the degree of plasticity of clays. Soil Temperature. The sources of heat in the soil are three in number ; namely, the sun's rays, chemical changes, and the original heat of the earth's inte- rior. The principal source is the sun. The heat derived from chemical changes is not great, and, indeed, is not even worthy of consideration, except in soils verj' rich in organic matter ; and here the changes occur only in the presence of comparatively high temperature due to the action of the sun. The third source is constant and of much importance. The soil temperature is influenced by a number of conditions, including exposure, atmospheric temperature, color, compactness, com- position, and moisture. Xaturally, the surfaces exjiosed to the greatest amount of sunshine get more heat than others. The nearer the angle of incidence of the sun's rays approaches a right angle, that is, the more perpendicularly the rays strike, the greater the amount of heat received. , The rapidity with which soils are affiected in either direction by changes in atmospheric; temperature varies widely, but with any soil it is only in the very upjiermost layers, the very surflice in fact, that any immediate corresponding rise or fall is obsers'cd. Great sudden changes affect the soil below the surface very slowly, and in the deeper layers the maximum and minimum temperatures occur nuich later than in the atmosphere above. The annual variation diminishes as the distance from the surface increases ; at fifteen feet the amplittide is, as a rule, less than 10 degrees F., and between fifty and eighty feet the tempera- tiu-e is constant the year round. The color of a soil exerts an important influence in the determination of its temperature. As is well known, a black surface exposed to the sun absorbs the heat rays more than a white one. A common illustration of this fact is the greater rapidity with which snow melts when its surface is dotted over with dirt and soot than when it is clean and white, owing to the absorption of heat by the dark ]>articles and its communication by conduction to the snow beneath and al)OMt. In the same way, soot and cinders work their way dt)wnward into the ice on a pond. Another illustration is the greater feeling of SOIL TEMPERATURE. 281 Avarmth conferred by black clothes than by white in the bright sun- shine. So, other conditions being the same, a dark soil is warmer than a light one, which reflects the heat rays instead of absorbing them. Observation has shown a diiference of more than 25 degrees F. in the temperature of black and white sands exposed side by side to the direct rays of the sun, but the white sand by reason of reflecting the heat rays will appear to be much hotter than it really is. The influence of compactness on soil temperature varies with the season. According to King,^ the general tendency of rolling the land is to make it warmer during bright, sunny weather, but in cloudy or cold weather it tends to promote cooling. He has observed that, at the depth of 1.5 inches below the surface, a rolled field may have a temperature 10 degrees F. higher than a similar soil not rolled, and at double the distance he has noted a difference of 6.5 degrees. This is due chiefly to the fact that a compact soil is a better conductor of heat than one containing large interstices filled with air. The character of the mineral and organic constituents of the soil and the amount of its content of water exert the very greatest influence upon its temperature. Hocks, sands, and mineral substances in gen- eral are better heat conductors than water, organic matter, and air, and they differ also one from another in conductivity. Organic matter is a particularly poor conductor of heat, and hence the greater the amount of humus a soil contains, the slower its response to the action of the sun. The great influence of moisture on soil temperature is due to the high specific heat of water, and to the loss of heat which accompanies the process of evaporation. The specific heat of ordinary dry soils varies from a fifth to a fourth of that of water, although in exceptional cases it may amount to nearly a half ; and the wetter the soil is, the higher will be the specific heat of the mass, that is, the greater the number of heat units necessary to warm a given weight 1 degree. Thus it happens that a light-colored dry soil may, in spite of the great influence of color, attain a much greater degree of warmth than a dark one which is damp. The different soil constituents have different specific heats, ranging from about 0.16 for certain sands and clays, to about 0.44 for dry humus, that of water being unity. Thus, to raise the temperature of 100 pounds of water 1 degree will require 100 units of heat, while to perform the same office for equal weights of dry sand, weathered porphyry, weathered granite, and humus, will require respectively 16, 20, 30, and 44 units. Therefore, the same amount of heat necessary to raise a given weight of water 1 degree will raise the equivalent weights of these substances respectively 6.67, 5.00, 3.33, and 2.27 degrees. But although the high specific heat of water is of importance in determining soil temperatures, the chief influence of moisture in this direction is due to the great loss of heat which accompanies the process 1 The Soil, New York, 1898. 282 THE SOIL. of evaporation, for the change from the liquid to the gaseous form is accomplished only at the expense of heat. The greater the amount of water evaporated from a given soil, therefore, the greater the expendi- ture of heat and the greater the lowering of the soil temperature. Conversely, the drier the soil, the less the evaporation, and the greater its Avarmth. Water does not, however, always tend to produce lower- ing of the temperature, for, in point of fact, it may and often does have the o])posite effect. In the spring, for instance, when the frost is not yet out of the ground and when the interstices are filled with cold water derived from the melting ice and snow, the warmer rain hastens the removal of frost, and, as it sinks into the soil, displaces downward the colder water and consequently raises the temperature. Changes in the Character of Soils Due to Chemical and Biological Agencies. Chemical action is constantly at work in the soil, not alone on the organic constituents, but upon the mineral matters as well. The changes which occur in the latter are of importance to the hygienist almost solely in so far as they affect the quality of the drinking-water. Complicated processes involving the decomposition of organic matters give rise to quantities of carbon dioxide which, being taken into solu- tion by the water in the interstices, assists in the production of still more complicated processes which engage the mineral constituents. The changes which, from a public health ])oint of view, are of the greatest interest are those which are in progress in the process known as the " self-purification " of soils, in which the complex organic mat- ters are broken up and reduced to simple chemical substances through the intervention of bacterial life. In the end, the carbon is oxidized to CO2, and the nitrogen either is set free, or is combined with hydro- gen in the form of ammonia, or is oxidized to nitric acid and nitrates. The process requires tke presence of atmosjiheric air and of moist- ure not in excess, and is favored by temperatures between 53° and 131 ° F., the most favorable being 98°. It })roceeds most vigorously and perfectly nearest the surface, and virtually ceases at a depth of more than three feet, little or no action occurring in the subsoil beyond that de]>th. If too much organic filth and its attendant moisture are pres- ent, the soil becomes boggy and the changes cannot proceed. An influence of very great importance in its eifects on the physical and chemical characteristics of soils is that exerted by earth worms, which live chiefly on half-decayed leaves, which they drag into their burrows to be used as food and as linings and ]>lugs for the burrows as well. According to Charles Darwin,^ their castings contain 0.01 (S per cent, of ammonia, and the humus acids, which have been proved to play a very important \x\vi in the disintegration of various kinds of I'oeks, ap[)ear to be generated within their bodies. They swallow earth both in the process of excavating their burrows and for the ^ The Formation of Vesjetablc Mould throwo;h the Action of "Worms. SOIL-AIR. 283 nutriment which it may contain, and exert an important mechanical action on the soil oTaius, reducing their size by attrition within their gizzards. After filling themselves with earth, they soon come to the surface for the purpose of emptying themselves. "In many parts of England a weight of more than lU tons of dry earth annually passes through the bodies of worms, and is brought to the surface on each acre of land, so that the whole superficial bed of vegetable mould passes through their bodies in the course of every few years." From various data, Darwin calculated that the castings, spread out uniformly, would form, in the course of ten years, a layer varying from 0.83, in the case of a very poor soil, to 2.2 inches in ordinarily rich soils. Their mechanical action and that of ants, moles, and other burrowing animals have much to do with keeping soils open and friable. Soil-air. The air in the interstices of the soil differs from that of the atmos- phere mainly in its richness in carbon dioxide, which arises from the decomposition of organic matters. It is also poorer in oxygen, but by no means alwavs in a corresponding degree, and it is usually C[uite humid by reason of the presence of soil moisture. The amount of carbon dioxide varies very widely in different soils and at different depths of the same soil, and it fluctuates very consider- ably also under differing conditions at any given point in the same soil. Other conditions being the same, the amount is most marked in soils rich in organic matter undergoing decomposition-changes. In soils poor in this respect, the amount may be no greater than in the atmos- phere. Pettenkofer, for instance, found in the air of desert sand, which was devoid of organic matter, the same amount as was present in the air immediately above it. In ordinary soils, the amount increases with the distance from the surface, as has been shown by Fodor,^who made a great ntimber of analyses of air at different depths at a number of places, the observa- tions extending over several years. The average amounts found at depths of 1, 2, 3, and 4 meters, expressed in parts per 1,000, were as follows : Depth in meters. 1 2 3 4 Station 1 4.8 13.7 18.1 6.6 14.3 28.4 20.1 28 7 Station 2 . Station 3 36 5 The influence of season also was shown by him to be very considerable, the highest amounts occurring during the hot months, and the lowest in \^-int€r. The averages bv months are presented m the following table : 1 Boden unci "Wasser, Bi-aunscliweig, 1882. 284 THE SOIL. Month. Depth in meters. January . February . March . . April . . May . . . June . . . July . . . August . . September October . November December 12.6 12.2 11.8 14.9 16.1 21.5 22.8 20.7 19.3 15.0 13.8 12.6 25.0 24.8 24.7 25.2 27.2 29.2 35.9 32.6 31.4 29.4 26.5 25.8 Fig. 1"; These results are only such as might be expected when we consider that decompositiou of organic matters proceeds most vigorously within certain limits of high temperature. Fluctuations in the amount present at any given point are due to a immber of conditions which include rainfall, the action of the wiud^ the rise and fall of the subsoil-water, and differences in atmospheric pressure and temperature. Rainfall, l)y filling the superficial interstices of the soil with water,, interferes with the natural process of soil ventilation and causes an immediate accumulation of carbon dioxide, which, however, is shortly followed by a diminution due to absorption of the gas by the water, which thus acquires an increase in its power of attacking and dissolving the mineral constituents of the soil. Inasmuch as the bulk of the absorbed rainfall i& held by the upper strata of the soil, its influence is more marked there than at greater depths. As it sinks downward, however, in very wet weather, it drives the air before it, and causes its escape at points where its egress is not obstructed. The action of wind is exerted in two ways : by perflation and by aspiration. By blowing strongly across the surface of the soil, it aspirates the air in the upper layers and causes an upward movement in the air below, or it may suck it out at one mo- ment and take its place the next. Again, it may blow with such force against the surface as to drive the contained air downward before it, so that the interstices become filled with ordinary atmospheric air. The action is more marked in soils of ordi- nary coarseness of texture than in very open soils witli large interstices, which permit freer movement in the upper strata. This may readily be demonstrated by means oi" a simple experiment with the apparatus shown in Fig. 17. Here we have a glass cylinder, inside which is a glass tube extending from the bottom Apparatus to show action of wind on soil air. SOIL- A IE. 285 and bent over at the top so as to form a U, into which an amount of water sufficient to form a seal may be iutrockiced. If now we fill the intervening space up to the top with sand, and then direct against the surface of the latter a current of air bv means of a bellows or by blow- ing sharply through a tube of glass or other material, the whole volume of air in the interstices is set in motion, which is communicated to the air within the enclosed tube, so that the water in the U-shaped depres- sion is caused to oscillate. If the water completely fills- the short leg of the U, it may be forced over and caused to drip. If, however, instead of employing sand, we fill the cylinder with coarse gravel, the oscillation of the water will be either less noticeable or entirely absent, the air which enters at one point on the surface commuui- oating its motion only to that immediately adjacent in the upper part. The rise and fall of the water in the subsoil assist in the production of variations in the amount of carbon dioxide ; on the oue baud, by its rise, forcing the rich soil air upward and outward, and, on the other hand, by its fall, drawing the soil-air downward and causing its place in the upper strata to be filled with atmospheric air with lo^y content of the gas. Differences in temperature and barometric pressure have also been mentioned as exerting influence on the motion of the ground air. In spring and summer, the ground air is colder and denser ; and in autumn and winter, it is warmer and lighter than the air above. Hence in the former, it tends to remain stationary or to sink ; wliile in the latter, it rises and mingles with the atmosphere, which, under proper conditions, replaces it. Again, these changes may occur in both directions within the same space in twenty-four hours. For instance, at evening and at night the atmospheric air, being colder, enters the soil ; while by day, being warmer, its direction is reversed, and air is drawn up from below. Movement due to temperature differences is almost constant, since it is only rarely that the temperatures of the air and soil are in agree- ment. The influence of barometric pressure-changes is not very great ; with fall in pressure, the tendency is toward upward movement, and with rise, toward downward movement ; but Fodor found from the study of a large number of observations that the actual observable changes were insignificant. With the various influences at work causing movement of the soil- air in all directions, it is plain that the soil, especially if highly permeable, is endowed with a sort of respiratory function which keeps it more or less well ventilated. Formerly, it was believed by Pettenkofer and others of the " ^Munich School " that the amount of carbon dioxide in soil-air might serve as an index of the amount of impurity and of the rate at which the latter is decomposing, and that comparison of the amounts obtainable from different soils would serve to indicate their relative cleanliness. But such is not the case with soils equally permeable, owing to the influence exerted on soil ventilation by so many varying and conflicting causes. 286 THE SOIL. Indeed, it has been proved by Fodor that a permeable soil extensively contaminated by organic filth may yield less of this index of impurity than one far cleaner, but less susceptible to ventilating influences. Soil-water. The moisture contained in the soil may be designated in three dif- ferent ways, according to its position and the forces by which it is held in place ; namely, hygroscopic, ca]iillary, and gravitation. Hygroscopic water is that which adheres to the surfaces of the soil particles in the presence of air. A certain amount of moisture is con- densed upon the surface of most solid substances exposed to ordinary dampness, and it adheres with great tenacity. The amount of water so obtained differs, other conditions being the same, according to the nature of the soil, some soil constituents surpassing others in their power to attract it. Thus, soils rich in organic matter (humus) have a greater degree of hygroscopicity than others in which this constituent is })resent to a lesser extent. In some soils, the amount of hygroscopic water is very marked by reason of the large amount of organic matter, and because also of the large surfiicc area presented by the soil particles. Some idea of the tenacity with which this moisture is retained may be derived from the fact that air-dried soils which appear to be quite dry — the dust of country roads, for instance — may yield as much as a tenth of their weight of water on complete drying by ordinary labora- tory methods. Both the moisture absorbed from the aii* and the water held on the soil grains by surface attraction after a condition of dccidcnl wetness has been changed by the draining away of the rest, may be termed hygroscopic. The capillary moisture is that which is held within those spaces which have been spoken of as capillary in their nature. Under ordinary conditions, these are intermingled with spaces which may not be so designated and which contain air, and so the caj)illarv moisture does not ordinarily ecpial the pore-volume. The water in the capillary spaces may be that which is retained after thorough wetting from above, or it may have crept upward or laterally from points comjiletely satu- rated. Capillary movement occurs in all directions, but it is most marked from below upward to points where water is being withdrawn by evaporation or by the demands of growing vegetation. The height to which water may rise by virtue of this force depends upon the diameter of the spaces ; the smaller the diameter, the greater the rise. Jurin's law of capillary movement is, that the height of ascent of one and the same liquid in a capillary tube is inversely as the diameter of the tube. Thus, water will ascend ten times as high in a tube having a diameter of 0.1 mm. as it will in another with a diameter of 1.0 mm. It follows, therefore, that capillary movement is most marked in soils of fine texture. C^apillary movement is influenced materially also by tcm])eratnre and bv the nature of substances held in solution. It diminishes as the SOIL- WATER. 287 temperature rises, and increases as the temperature falls, so that cooling a soil uniformly will cause increased capillary movement, and heating it will cause a fall. But with uneven temperatures, the motion will be different according as the temperatures vary. Thus, if the lower part of a column of soil be cooled, the surface tension of its contained water will be increased at that point, and water will be attracted from the parts above, gravity assisting ; whereas, if it be heated, its con- tained water will be attracted upward. In saturated soils, motion of the water in any direction is influenced very greatly by temperature, because of the effect of heat in changing the viscosity of water. The higher the temperature, the greater the diminution in viscosity and the freer the movement. The influence of dissolved substances depends upon their nature, some favoring, and others retarding, movement. The rate is increased by the presence of nitrates, and is diminished by common salt and sulphate of calcium ; but the favoring influence of the presence of nitrates is counteracted most markedly by organic substances produced in the decomposition of matters of vegetable origin, for a minute trace of these completely neutralizes the effect of such amounts of the former as are commonly present in cultivated soil. It is self-evident that anything tending to the diminution of capillarity of a soil diminishes the rate of capillary flow. When the soil is worked in such a way, therefore, as to produce an open, crumbly condition in place of one of compactness, the rate of capillary move- ment within it is diminished very greatly. We come now to the third division, w^hich has been designated as gravitation-water. This is the water which has drained away through the soil by the force of gravity and accumulated in the subsoil over an impermeable stratum which has arrested its farther downward journey. This is what is commonly known as ground-water, or subsoil-water. Its zone extends from the surface of the impermeable barrier upward to that point where the interstices of the soil cease to be completely filled with water, but are filled partly wdth air. This point is knoAvn as the ground-water level. The zone above it, through which water is moved in the capillary spaces, is known as that of the capillary soil- water, and extends as far as the water is moved through that force. Above this, at and near the surface, is the zone of evaporation, from which water is evaporated into the atmosphere. The impermeable stratum beneath the subsoil-water may be either very fine sand, compact clay, or rock. It may be thin or thick, accord- ing to circumstances. Below it, there may be a succession of alter- nating permeable and impermeable strata, so that in driving deep wells a variety of strata are pierced, and waters of varying composition may be secured. Dense clay is practically impermeable to water, but at the same time it can communicate its moisture to surfaces with which it comes in immediate contact, a fact which renders necessary the in- terposition of damp-proof material in the foundations of houses built upon it. 288 THE SOIL. Rocks vary greatly in impermeability ; the densest of them contain very .small amounts of moisture in their pores, while others are so porous that they may contain as much as a third of their volume of water. Again, most rock deposits are more or less fissured and seamed, and thus permit to a greater or less degree the passage of water at these points. The water-bearing stratum is usually gravel or sand, but may be porous or fissured rock. Its depth is exceedingly variable, depending upon local geological conditions, and at two points not widely separated, it may be respectively slight and considerable. The ground-Avater is in constant motion both laterally and vertically. Its lateral movement, whatever its rate, depends upon the configuration of the impermeable layer below, and not upon that of the surface of the land. Generally speaking, the direction of the movement is toward the nearest large body of water, be this the sea, a lake, or a river ; but it is not often possible to determine, except in a general way, from surface observations, whether at any given point the flow is in one direction or another. This is especially true when the water-bearing stratum is thin and underlaid by an impermeable stratum of veiy irreg- ular conformation. The rate of movement is also exceedingly variable ; it may be fast, or slow, or hardly perceptible. In ^Munich, for instance, according to Pettenkofer, it amounts to about fifteen feet daily, while at Berlin, it is only very slight, and at times is wanting. It is influenced by the configuration of the subsurface, by the perme- ability of the subsoil, by the amount of the accession of moisture from rainfall and melting snow, by the obstacles interposed by the roots of trees and other plants, by others at its outiall, and by the withdrawal'of moisture by the needs of vegetation and of communities of men. The rise and fall of the ground-water — that is, its vertical move- ment — depend chiefly U])on the amount of rainfall ; and, on the other hand, upon the rate of withdrawal by evapcjration, vegetation, and water supply of communities, and upon the freedom of, or obstacles to, the outflow. The effect of rainfall is generally not immediately perceptible, for so much time intervenes between heavy falls and penetration that a falling of the ground-water level may continue to be observed for a long time after a period of great wetness ; but when the level rises, it is a ]iroof that additions have been received from above, though per- haps the accession has travelled through a long chstance in the soil. When the level fiills, it is a sign that the upper strata have become dry through evaporation, and that ca})illarv attraction has carried moisture upward to replace the loss. The rise and fill of the ground- water level may be determined by measuring from day to day the distance between the surface of the soil and the height of the water in a inmil)er of wells in a given locality. This may readily be done by means of a tape-measure or chain to which SOIL-WATER. 289 is attached a rod bearing a nmnber of shallow metallic cups which are lowered into the water. The distance between the point on the chain at the mouth of the well and the uppermost cup in which water is found indicates the position of the water-level with respect to the surface. By removing obstacles to the outfall of the underground river as it sometimes is called, and by creating new outfalls by ditching more or less deeply, according to individual conditions, by sinking drainage wells, or by laying drain tile beneath the surface at such depths as may appear to be advisable, the level of the ground-water may be con- siderably lowered, and the soil thereby rendered correspondingly drier, and also, by reason of the influence of water on soil temperature, warmer. Sources of Soil-water. — The principal source of soil-water, it is hardly necessary to say, is the rainfall, but by no means all of the water precipitated from the atmosphere during a storm penetrates to the subsoil. Light rains may be wholly lost by evaporation, and heavier ones, especially during active vegetation, may penetrate but very slightly beneath the surface. In early spring and in au- tumn, the amount which percolates downward is naturally much larger in proportion. A by no means insignificant amount of moist- ure is that derived by absorption and condensation from a moist atmos- phere. In periods of drought, this power of dry soil to absorb water from humid air is of the greatest value to vegetation. The amount absorbed differs according to the nature and hygroscopicity of the soil elements. Thus, a soil rich in humus will attract more water than another composed wholly of sand. Condensation of water occurs when the surface is cold and in contact with moist air. This condensation may occur from above or from the rising moist soil air just below. A third source of moisture, of no great importance, is the breaking up of organic matter into its constituent elements, in which process the hydrogen is in great part ultimately released in combination with oxygen as water. Another and exceedingly important source of soil moisture, important not because of the amount, but because of the quality of the water, and because of its possible effect on the supply of drinking-water and on public health, is the waste waters incident to human life, which in so great a proportion of communities are dis- charged directly into the soil, where, being out of sight, they are equally out of mind. Loss of Soil Moisture by Evaporation. — The amount of water which a soil loses by evaporation is influenced by a number of factors, which include the water content of the soil, the height of the permeable layer, the composition and structure of the soil, and the character of its surface, and, particularly, whether it is covered. In other words^ the rapidity of the process is proportional to the combined area of sur- faces exposed, and to the facility for replacing the loss by withdrawals from below. 290 THE SOIL. Influence of Vegetation on Soil Moisture. — The amount of water in soils is infiuenced greatly by growing vegetation, which requires a vast supply for the proper maintenance of its functions. It withdraws it by absorption by the roots, which extend downward to surprising depths, the roots of wheat, for instance, attaining sometimes a length of eight feet and more. From the roots, the water passes into the cir- culation of the plant, assists in the various physiological processes, and then, for the most part, is given off from the leaves into the atmos- phere. It has been calculated by Pettenkofer that an oak evaporates more than eight times the rainfall, and that the Eucah/ptas globulus is even more active. The difference between the rainfall's contribution and the amount exhaled represents the amount which has been with- drawn by the roots from the capillary spaces and from the water-table itself. As the water in the capillaries is relinquished by them, more comes up from below to take its place. Thus it is that a plant or tree acts duriug tlie growing season as a constantly workiug suction appara- tus tending to dry the ground, and so may be explained, in part at least, the condition of wetness that is acquired by some lands after removal of trees. All growiug crops withdraw enormous amounts of water, and after the growth becomes well advauced, it is the capillary water upon which dependence is placed, for the rainfall penetrates but a short distance into cultivated land, and most of it is lost by evaporation. Were it not for the capillary water supply, no crops could be raised, except under most extraordinary conditions of Avcather and by artificial irri- gation, since but a short period of drought would suffice to produce wilting. According to Stockbridge,^ " The quantity of water thus required and evaporated by different agricultural plants duriug the period of growth has been found to be as follows : 1 acre of wheat exhales 409,832 pounds of water. 1 " " chiver exhales 1,09G,234 " " " 1 " " sunflowers exhales 12,58o,994 " " " 1 " " cabbage exhales 5,049,194 " " " 1 " " grape-vines exhales 730,733 *' " " 1 " " hops exhales 4,445,021 " " " But the influence of vegetation on the water content of the soil is not limited simply to its withdrawal and evaporation into the atmos- phere, for it acts in the other direction to impede surface flow and sub- surface drainage. This is seen more particidarly in the case of trees and forests. The forest cover keeps the soil granular and promotes downward percolation ; the tree roots, penetrating in all directions, present an effective obstacle to rapid lateral movement through the soil. Removal of forests and clearing away the surface of the forest litter promote sudden and destructive freshets in the s})ringtime and drought when, later in the year, the water is needed. The ill effects of deforest- ation are noticed particularly in parts of Maine and in the Adiron- dacks, where streams that formerly ran full .the year round are raging » Rocks and fSoils. New York, 1888. SOIL- WATER. 291 torrents when the winter's snows are melting and but insignificant brooks or wholly dry during the smiimer months. It has been stated by Major Raymond, of the U. S. Engineers, that, in forest areas, four- fifths of the rainfall are saved, while in cleared land the same amount is lost by evaporation and surface flow. Other Effects of Vegetation Upon the Soil. — In addition to its influence on the movement of soil -water and on its amount, vegeta- tion is an important factor in the determination of soil temperature and of the amount of mineral matter available for succeeding growths. The deeply penetrating roots bring to the tissues of the grooving plants a large amount of mineral matters from the subsoil. On the death and decay of the plant, these matters are returned to the soil at its surface, wdiere they are available for reabsorption as plant food. The effect of vegetation on soil temperature is of much importance in both hot and cold climates. A barren soil or one from which veg- etation has been stripped absorbs the heat rays of the sun more rap- idly and becomes much hotter than one which is protected by growth, of any kind. The air above the soil becomes hotter, too, because of greater heat radiation, and the difference in the surface temperature of bare ground and that covered by grass or other vegetation is furtlier increased by the cooling eflPect of evaporation of moisture from the leaves. Herbage acts as a protection against excessive heating in hot climates, and as a blanket to prevent loss of heat in cold ones. In summer, the areas covered by vegetation are cooler than those which are unprotected against the direct rays of the sun, and in winter, they are warmer because of the obstacle to heat loss. Trees obstruct the sun's rays and impede wind currents, and thus, the soil is cooler and at the same time suffers less loss of moisture by evaporation. The obstruction of the wind currents deprives the soil air of one of the influences having to do with its movement, and thus interferes with soil respiration. The obstacle opposed by trees to the motion of air is so great that, in the interior of a piece of woods, the air may be quite calm while a gale is blowing outside. In winter, the obstruction of the sun's rays aids in the conservation of the soil heat by preventing the accumulated snow from melting, and thus keeps the surface protected by a blanket. In cold climates the influence of trees may be at the same time per- nicious and beneficial ; that is to say, pernicious, in that the ground is colder and moister than it would be had the sun's rays free access, and beneficial, in that the trees afford protection against wind. The judicious removal of trees will often render a climate more equable. In hot climates, as in cold, trees should be removed only in case of necessity and after due consideration of the probable results. The hottest spots in hot countries are those deprived of the beneficial influ- ences of vegetation. It may not be out of place here to mention the supposed agency of woodland in protecting communities from " malarial exhalations " from swamp localities. That the interposition of a belt of trees has been 292 THE SOIL. followed in a number of instances by decided improvement in public health so far as malaria is concerned, cannot well be denied ; but the improvement is not owing to the fancied property of leaves to con- dense upon their surfaces the malarial poison, but to the fact that the winged bearers of this poison, blown along by the wind, are filtered out of the air by the leaves, or themselves seek the protection thus afforded against farther involuntary- movements, and attach themselves to the leeward side of leaves and trunks. Pollution of the Soil. The soil receives polluting matters of infinite variety and in Avidely differing amounts, but their nature and their amount are of less importance relatively than their point of entrance. Some of these pollutions are unavoidable, and these, indeed, are the ones concerning which we may give ourselves the least concern ; others are avoidable, though not always, or even usually, without the incurring of e.\])ense. The unavoidable pollutions include the urine and droppings of animals, the carcasses of such as have died and have escaped the notice of other animals that act as scavengers, and vegetable matters of every conceivable kind in various stages of decay. Excepting under very unusual conditions, such, for instance, as may exist in time of war or flood or epidemics, when large numbers of horses, cattle, and other animals are killed or die, these, lying at or near the surface, are of comparative unimportance, since, exposed to natural processes of jjurification, they are resolved into simple innocuous substances, which are absorbed by plant life or washed downward into the soil. The ayoidable pollutions are mainly those which man deposits beneath (the surface, and these are first, and of minor importance, the bodies of tlie dead, and second, of vast importance, the excreta and other organic filth that constitute sewage. The temporary storage of filth in water-tight receptacles built under ground can, of course, do no harm to the surrounding soil, but it is not into such that man usu- ally chooses to deposit his waste. Water-tight cesspools gradually become filled and then require to be emptied, while tliose with pervious bottoms ])ermit the escape of the contents downward, require no thought or care, and are, therefore, a source of contentment and of saving of expenditure. The filth thus introduced is, however, below the zone of bacterial activity of the beneficent kind, and becomes stored up in the subsoil or is washen inches. After 130 days, the bacilli on the surface had multi|)lied, and where they had been placed eighteen inches below, they could also be found in the surface layer. Later on, in the winter, no results could be obtained ; but in the spring, he moistened the patches with sterile bouillon in very dilute conditi(^n, and afterward succeeded in obtain- ing growths. ^ Munchener medicinische Wocliensclirift, May 2, 1899. ^ British Medical Journal, .Tan. 8, 1898. SOIL AND DISEASE. 299 This positive result accords with the views of Germano/ who found that typhoid bacilli will live for months when incompletely dry ; but according to Flligge, they do not survive complete drying longer than fifteen days. In air-dried condition they appear to have unimpaired vitality for some days, according to Brownlee/ who dried and sterilized ordinary soil and then infected it with a broth culture of typhoid and kept it at 98° F. for a day. It was then left exposed to the air for a week, during which time it became sufficiently dry to be easily scat- tered by the breath. Cultures from this gave positive results. But it should be remembered that air-dried soil contains considerable hygroscopic water ; consequently his bacilli were doubtless fairly well supplied with the necessary moisture. Of more importance, ap- parently, than the question of moisture — for all soils possess some — is the nature of the contained organic matter. Dr. Sidney Martin^ has shown that unpolluted (virgin) soils are inimical to the typhoid bacillus, regardless of the amount of their contained organic matter of vegetable origin, while specimens containing polluting material of animal origin favor its existence. Such, after sterilization, were planted successfully, and it was learned that, in the presence of moisture, differences in tem- perature had but little influence. Thus, the organism thrived about equally well when specimens were kept at 98° F., at ordinary room temperature, and as low as 37° F. By no means the least interesting observation made was with regard to the duration of viability of the bacillus. In one of the sterilized polluted soils, the organism was still active at the expiration of 456 clays ; and even then, after thorough drying and pulverization, active growth could be obtained. In com- pany with various species of bacteria, among which the predominant kinds were members of the B. coll group, it was recovered after 50 days' exposure to temperatures ranging between 37° and 61° F. Later experiments,* in which the typhoid organism was planted with different soil bacteria, proved that various species from a particular soil had the power of completely exterminating it within a short time, while others had no influence whatever. Therefore, it would appear, whether or not the typhoid organism can exist in a given soil, will depend upon the kinds of soil bacteria present, as well as upon special conditions of temperature and dampness. Dr. Martin found the period of vitality in unsterilized soils to be about 12 days, but in no case did the organism appear to multiply. He alleges two reasons why one cannot expect it to thrive in the surface layers of soil. The first is that the more hardy colon bacillus is most commonly not to be found, excepting in cases of recent animal pollution, or present in sparse proportion only ; and this would indicate that the surface soil is unfavorable to the growth and vitality of non-sporing bacteria of intestinal origin. The second reason is the frequent presence of physical conditions inimical to microbial life, ^ Zeitschrift fiir Hygiene und Infectionskrankheiten, XXIV., p. 403. 2 Public Health, .Jan., 1899, p. 272. ^ Report of Local Government Board, 1898, London, 1899. * Ibidem, 1900, London, 1901. 300 THE SOIL. suggested by the enormous preponderance of spores of aerobic V)acteria. The eiFect of temperature changes due to the presence of animal ex- creta mixed ^vith the soil is sho^^^l by Gaertner^ to be considerable. He introduced cultures of various organisms in -^"ire baskets into the interior of compost heaps of various composition, which became heated ti> different extents, and observed that the bacilli of typhoid and cholera were the least resistant of all. With rapid and marked heat- ing, their life was short ; but it appears probable that in the absence of heat, even with the given surroundings, they may live through the winter. Under the ordinary heating that occuiTed in the compost, these two organisms were destroyed in a week, while the bacillus of tuberculosis remained virulent a nimiber of months. But aside from what we glean from scientific research with the spe- cific organisms, we know from experience that there are many places with polluted soils where typhoid fever was unknown until the impor- tation of a single case from without, and that, afterward, sporadic cases, for which no convmcing explanation is afforded, have occurred at vaiying interv'als. And in countr^' districts, whose inhabitants are not given to travelling much beyond the confines of their farms, it is noticed frequeutly that single cases occur in the same household at intervals of a year or longer. In such cases, it seems hardly reasonable to say that the original case has left nothing as the exciting cause for later attacks, and that fresh introductions of the specific organism must have occurred from some unknown source, for it is not unlikely that the variety of condi- tions that affect the viability of the org-anism may, in some cases, act to keep it alive, and, on occasions, stimulate it into a condition of aug- mented activity. Cholera. — Concerning the relation of this disease to the soil, there is but little to be said. Prior to the discovery of the specific organism, the soil theory of the origin of epidemic outbreaks had considerable vogue ; but now it is known that, even in times of greatest prevalence of the disease, the organism has never been found under natural con- ditions in the soil. It can be kept alive under certain favorable con- ditions of moisture and heat for vaiying periods ; but imder natural conditions it is one of the least resistant bacteria and quickly dies. We have no evidence whatever that cholera is a soil disease. Bubonic Plague. — This has been regarded as a soil disease ; and it has been believed, from the fact that rats have been conspicuous as vic- tims of it in the early stages of its devastating outbreaks, that these animals have ac'quired the infection in the soil, and have brought it to the surface, and thus acted as its carriers. But rats are notorious as frequenters of places where filth of all kmds accumulates, and it is not strange that where they and filth abound, they become diseased, if the infective agent is })resent. TIk' tremendous epidemics that have raged within the past few years ' Zeitselirilt fiir Hygiene und Infectionskninkheiten, XXVIII., p. 1. SOIL AND DISEASE. 301 presented unusual opportunities for extensive study, which thus far has afforded no evidence whatever that the disease is soil-borne. It has been shown, on the contrary, that the plague bacillus does not long survive in the soil. Yokote,^ for instance, obtained a number of mice that had died of the disease, and buried them in garden soil which was kept well moistened. At short intervals, he removed them successively, and examined them culturally and by animal experimentation. The longest interval between interment and proof of continued virulence was thirty days, and it was shown that the higher the temperature and the more rapid the decomposition, the shorter the life of the organism. The soil in the immediate vicinity of the animals proved to be free from the bacilli. Diphtheria. — Although there is no proof that the bacillus of diph- theria is found even as an occasional lodger in the soil, there is a gen- eral agreement that a close connection exists between soil dampness and the prevalence of this disease. It is true that experiment has demon- strated the viability of the organism in moist soils for limited periods, but it has never been found in soils other than those in which it was deposited intentionally. The common belief is that a moist soil is an invariable concomitant of unusual prevalence, and that in times of comparative freedom from the disease, the soil is dry and the level of the ground- water low. "As long as the soil is well washed by the winter's high tide and afterwards dried and aerated during the summer's low tide, all goes well : but so soon as these salutary movements are arrested or their order disturbed, diphtheria prevails, reaching its acme of prevalence when stagnation at a relatively high level is most complete." ^ According to Dr. S. M. Copeman,^ there appears to be no direct relation betAveen epidemics and rise or fall of the ground-water, " pro- vided that the structure and atmosphere of the houses are not affected. Many districts, which, usually dry, are liable to occasional floods, are remarkably free from the disease, so that it appears that a persistent impregnation of the soil with moisture is of more importance than fluctuations in the height of the ground-water, particularly if these have any considerable range." Opposed to the views above expressed are the conclusions based on a most careful and extensive investigation by Dr. Arthur jSTewsholme,* of epidemics of diphtheria in all civilized countries and their incident conditions of rainfall and soil moisture. Dr. Newsholme's eminence as a skilful interpreter of the value of statistics, and the fact that no such exhaustive inquiry into this question has ever before been made, entitle his conclusions to more than ordinary weight. Admitting that personal infection is the chief means by which diphtheria is spread from town to town, and from country to country, he summarizes his obser- ' Centralblatt fiir Bakteriologie, Abth. I., XXIII., p. 1030. ^ Notter and Firth, Treatise on Hygiene, 1896, p. 463. ^ Stevenson and Murphy, Treatise on Hygiene, 1892, Vol. I., p. 338. * The Origin and Spread of Pandemic Diphtheria, London, 1898. 302 THE SOIL. vations on the relation between rainfall and ground-water and the origin of epidemic diphtheria as follows : " 1. An epidemic of diphtheria never originates, in the towns and countries in which I have been able to collect facts, Avhen there has been a series of years in which each year's rainfall is above the average amount. " 2. An epidemic of diphtheria never originates or continues in a wet year (/. e., a year in Avhich the total annual rainfall is materially above the average amount), luiless this wet year follows on two or more dry years immediately preceding it. " 3. The epidemics of diphtheria, for which accurate data are avail- able, have all originated in dry years (/. e., years in which the total annual rninlall is materially below the average amount). " 4. The greatest and most extensive epidemics of diphtheria have occurred when there have been four or five consecutive dry years, the epidemic sometimes starting near the beginning of this series, at other times not until near its end. " 5. Dry years imply low ground-water, and we find, therefore, in the years of epidemic diplithoria that the ground-water is exce]itionally low. The exact variations in the ground-water which most favor epi- demic diphtheria cannot, with the data to hand, as yet be stated ; but it is probable that when this is cleared up it will become clear why in exceptional years which have a deficient rainfall epidemic dijihtheria is either absent or but slight." It has often been pointed out that local soil conditions causing dampness of habitations even in dry years, such dampness, for instance, as obtains in houses built over wet im]>ervious clays, conduce to out- breaks of diphtheria in the dwellers therein ; but, as is well known, such dampness acts as a very imj>t)rtant depressant of the vital forces, and prepares the raucous membranes of the respiratory tract for the favor- able reception of specific organisms of various kinds. Malaria. — It has ever been held that the most intimate relation exists ])ctween the soil and malaria, especially prominent in districts abounding in marsh lands. It has been noticed repeatedly that in malarious comitries the upturning and excavation of wet or damp soil are commonly followed by the occurrence of the disease among the laborers so engaged. Infection appears to be especially common after sundown and at night ; wherefore dwellers in places where the disease is rife have been enjoined to avoid going about after nightfall. This is explained by the assam]ition that the poison, resident in the soil, is brought up into the atmosphere by the soil air, Avhich rises toA\arenings. While it may be jwssible that the soil offers the cause — a residence Avithin itself Avhere vitality may be conserved and, perhaps, multiplication favored — we have as yet no evidence bearing on the question. Affainst the theorv of causal relation between disturbance of the soil and outbreaks of yellow fever, many instances may be cited, in which, under apparently favorable conditions, the occurrence of the one has not been followed by the other. Thus, in 1894, the streets of Jack- sonville, Fla., and of Brunswick, Gra., were dug up during the entire summer; in 1899, at Mobile, Ala., about 125 miles of aqueduct and sewers were laid during the summer; between 1890 and 1900 the streets of New Orleans, I^a., were extensively dug up in the course of public improvements, and in not one of these places did the public health appear to suffer in consequence of the excavations. On the contrary, in Xew Orleans, when it was visited in 1897 and 1898 by yellow fever, the parts of the city where the streets were dug up hap- pened to be singularly free from the disease. As in the case of malaria, the soil theory is giving way to that of transmission by mosquitoes. Other Diseases. — With regard to the connection which may exist between the soil and dysentery, scarlet fever, and other diseases, con- EXAMINATION OF SOILS. 307 cerning the organisms of which we are in the dark, it is best to admit frankly that we do not know, rather than to make general statements based on imaginings. Our actual knowledge of the relation of the soil to disease amounts in brief to this : that surface dampness is favorable to the develop- ment of certain diseases, as rheumatism, neuralgias, and affections of the respiratory tract ; that concerning some diseases there is a wealth of purely circumstantial evidence of connection, opposed in each case by evidence to the contrary ; that the soil is the home of many species of organisms, some of which are pathogenic, and offers under certain favorable conditions an at least temporary asylum to others ; but the preponderance of evidence thus far goes to show that under normal conditions the soil is more likely to prove hostile than hospitable to most of the infective agents with which we are well acquainted. Examination of Soils. The complete examination of a soil includes chemical, physical, and bacteriological determinations, but inasmuch as the chemical analysis, beyond the estimation of water and organic matter, is of no especial interest to the sanitarian, though of great importance to the agricul- tural chemist, we shall, with the exceptions noted, confine ourselves to the processes involved in the physical and bacteriological tests. In taking samples, a place should be selected which faii"ly represents the locality, and under some circumstances a number of specimens should be obtained. These may or may not be mixed and treated as one. About two pounds of the soil may be broken up by being passed through a coarse sieve, then spread out and left for one or several days exposed to the air, and to that extent dried. To deter- mine the relative proportion of the grains of different sizes, a weighed amount of the sample is now passed through a series of sieves of vary- ing coarseness, made of metal or porcelain with circular open spaces, which in each sieve are of uniform diameter. Those used by the German scientists have openings respectively ^, 1, 2, 4, and 7 mm. in diameter, by means of which a specimen is separated into grains of less than I, from |^ to 1, from 1 to 2, from 2 to 4, from 4 to 7, and over 7 ram. in diameter. Other sized openings may be used, but these fulfil all requirements. The specimen is passed first through the coarsest of the set, and then, in order, down to the finest. If the particles adhere firmly, the separation is done best with the assistance of water ; and should it be necessary, a pestle covered at the working end with rub- ber may also be employed. The separate parts are then dried and weighed, and their respective amounts expressed in percentages of the whole. The finest particles, that is, those of less than J mm., may be separated still further by the process of washing in an elutriating apparatus, of which there are several kinds, none of which, however, gives results that are more than approximately accurate, since so many different 308 THE SOIL. forces and conditions come into play to influence the process. With some, the separation is effected by causing the particles to settle down- ward through a volume of water, the heaviest ones reaching (theoret- ically, but not wholly in practice) the bottom first, and the lightest settling out last or remaining a long time in suspension. An apparatus of this sort is shown in Fig. 18, which requires no explanation. Another, known as Knop's silt cylinder, is shown in Fig. 19. This is a cylinder carrying lateral tubes fitted with stopcocks, situated at equal distances (10 cm.) apart. The sample is placed in the cylinder, which is then filled with water and well shaken. After a given time the upper stopcock is opened and the water above it is drawn off. Then after the lapse of another interval, the second is opened, Pig. 18. Fig. 19. ==^ 6t=a^ -iV ^^ Apparatus for separation of fine particles of soil. Knop's silt cylinder. and next, in the same way, the third. The process is repeated until the wash water comes away clear, then the lowest tube is opened, and the rest of the water above the remaining material drawn off. The different portions may then be collected, dried, and weighed, and their relative proportions expressed as before in percentages. Or the residue may be dried and weighed and the remainder estimated by difference. By another method, the washing is carried out by me^ns of an up- ward flow of water in a conical vessel, at the bottom of which the sample is placed. The water, delivered through a tube reaching to near the bottom of the vessel, carries the lighter finer particles upward and out through the exit tube near the top. Such an apparatus, known as Schultz's, is shown in Fig. 20. Pore-volume. — The pore-volume is determined very simjily by addincr to a volume of water in a gfraduated cylinder a known volume EXAMINATION OF SOILS. 309 Fig. 20. Schultz's elutriating apparatus. of soil in the dry state, and noting the height to which the water rises. If, for instance, to a liter jar containing water up to the 500 cc. mark, we add 500 cc. of dried soil in as nearly as possible its natural state of compactness, and observe that the level of the water is in consequence raised to the 850 cc. mark, it follows that the increase, 350 cc, represents the actual bulk of the soil grains, and that the difference between this and the volume occupied originally by the sample (500 cc), that is to say, 150 cc, represents the amount of interstitial space filled with air. Then, since 500 cc of soil contains 150 cc. of air space, it follows that the pore-volume of the sample is x in the equation 500 : 150 : : 100 : a;, or 30 per cent. In order to approximate more closely the natural condition of compactness, the sample may be taken from the soil by means of a metallic cylinder with a cutting edge. It is then dried in order to expel the contained water, which otherwise would constitute a source of error, and is then added to the water in the liter jar as before. Permeability to Air. — Permeability to air may be determined by forcing measured volumes of air under constant pressure through a cylinder closely packed with the sample, and noting the amount which is delivered during any given unit of time. In making comparison tests between different soils, the same conditions must be observed in every case ; that is to say, the length of the column of soil in the cyl- inder, the pressure employed, and the unit of time. A still farther condition which should be observed, but which is commonly disre- garded, is the temperature of the air, for, as is the case with liquids, the viscosity of gases varies with changes in temperature, though not in the same direction. The viscosity of liquids is increased with dimin- ished temperature, whereas in the case of gases the reverse is true. Disregard of this fact leads to important degrees of error. The apparatus for this determination, showm in Fig. 21, comprises a gas-holder (A), a gas-meter (_B), and a cylinder (C) provided with a manometer (D). For the purpose of keeping the soil in position, tightly fitting perforated disks (E and J^) of metallic gauze are intro- duced into the cylinder at both ends of the column of soil. In the preparation of the cylinder, the disk F is first introduced, and then the soil is added a little at a time, and made as compact as pos- sible by striking the lower end of the cylinder downward with reason- able force against the table. When the desired length of column has been reached, the disk E is introduced, between which and the inlet end (^G) an air space of sufficient size is left to insure uniform pressure 310 THE SOIL. against the entire surface of the disk. The inlet end is closed by means of a tightly fitting rubber stopper having two perforations, one of which carries the inlet tube from the gas-meter, and the other the manometer indicating the pressure employed. The pressure is obtained by means of a column of Mater communi- cating with the chamber of the gas-holder, which is connected by a rubber tube with the inlet of the meter ; and it is regulated by a screw Fig. 21. Apparatus for determination of permeability of soil to air. pinchcock on the outlet tube of the latter. The force is applied, the reading of the meter is noted, and at the expiration of the unit of time, one, five, or whatever number of minutes it may be, the reading of the meter is taken again. Permeability to Water. — The permeability of a soil to water is expressed in terms indicating the amount of water wliich will ])as« from above downward through a column of saturated soil during any EXAMINATION OF SOILS. 311 given unit of time under a given pressure. The apparatus for this determination, shown in Fig. 22, consists of a metallic cylinder (J.) with a perforated or gauze bottom on which the sample of soil is packed closely, and another cylinder {B), likewise of metal, provided with a number of outlet tubes (c), at regular intervals, preferably of 5 or 10 cm. The lower end of B fits tightly into the upper end of A, and the joint is made impervious to water by means of adhesive plaster, sealing-wax, or other suitable material. The soil within the lower cylinder is kept in place, and its surface kept intact, by means of a Fig. 22. Apparatus for determination of permeability of soil to water. superimposed disk of gauze or coarse cloth. The outlet tubes, provided with cocks, serve to maintain a constant level, and, therefore, a constant pressure of water as desired. Water is admitted in a constant stream to the cylinder through its upper end, by me^ns of a rubber tube con- nected with a water faucet. If it be desired to employ the highest pressure obtainable with the apparatus, all the corks of the ou^et tubes, except the upper one, are kept in place. In this case, the pressure would be expressed by the distance between the top of the soil under investigation and the uppermost outlet, through which the excess of 312 THE SOIL. water from the faucet is allowed to escape, by way of a rubber tube leading to a sink. Similarly, any other height and pressure may be employed by removing the cork of the corresponding outlet, which thus becomes the effluent. AYhatever the height maintained, it is necessary to keep the delivery end of the inlet tube below the surface of the water The process is as follows : Having chosen the pressure and adjusted the waste tube to the proper outlet, the water is allowed to run in and force its way down through the soil until the latter becomes saturated. In order to insure complete saturation, it is best, however, to immerse the soil cylinder, in order that all the air may thereby be displaced upward. When this has been accomplished and water begins to run or drip through the gauze bottom, the time is noted, and the discharged water is received in a suitable graduate. At the expiration of the unit of time, the latter is removed and its contents measured. The experi- ment may be repeated as often as may seem advisable, and the effects of varying pressures may also be determined. Water Capacity. — The power to hold water is determined by means of a metallic cylinder of known capacity with a gauze bottom. This is weighed, then iilled with the dried sample, and again weighed. The soil next is saturated completely by immersion of the cylinder in water, and then it is allowed to drain as long as water continues to escape. When the water ceases to drain away, the cylinder is wiped dry outside, and the weight of the whole is taken again. The increase in weight is the amount of water retained, and it may be stated in percentage of the pore-volume, which should have been determined previously. Capillarity. — The height to which water will rise in a column of soil by capillary attraction is detennined l)y packing the sample tightly into a graduated glass tube, the lower open end of which is covered with coarse linen tied securely on, so as not to slip. The tube is sup- ported with its cloth-covered end resting in a shallow dish filled with water, which is kept at constant level. The height to which the water rises through the column of soil is noted from time to time, until ascent ceases. The change in the color of the soil, due to wetting, indicates the progress of the action. Moisture. — The amount of moisture in a soil is determined most accurately by taking a sample in its natural condition, by means of a brass cylinder with a cutting edge, weighing a portion of it, and then drying it in an air bath at 105° C. imtil it ceases to lose weight. The difference between the original and final "weighings represents the amount of water in the given weio^ht of soil. If it is desired to know the amount of water which the same soil will absorb from a saturated atmosphere, the thoroughly dry sample may next be placed with a dish of water under a bell-glass. The confined air will become saturated with aqueous vapor in a short time, and this will be absorbed by the soil up to the limit of its capacity, which is shown when its weight no longer continues to increase. EXAMINATION OF SOILS. 313 The hygroscopic moisture of a soil may be determined roughly by air-drying a sample and then taking a known weight of it and heating it in an air-bath at 105° C. ; or by exposing it to a dry atmosphere in a bell-glass containing an open dish of concentrated sulphuric acid, until it ceases to lose weight. Organic and Volatile Matters. — Since it is impossible to deter- mine by ordinary processes the exact amount of organic matter present in any soil, it is necessary to designate the diminution in weight which occurs on subjecting a sample to such a heat as will burn oiF the organic matter, and which represents other losses than the latter, as " loss on ignition " or " organic and other volatile matter." For this determination, the soil which was used for the estimation of moisture, or another sample, thoroughly dried, may be placed in a platinum dish and heated over a Bunsen flame at no higher temperature than is suffi- cient to keep the dish at a dull-red heat. When all the organic matter has been destroyed, the residue is allowed to cool, and is then moistened with a little saturated solution of carbonate of ammonium, in order to restore the carbon dioxide that belongs to the inorganic constituents, then dried and gently ignited to expel the excess of ammonia, and finally weighed. The loss represents organic matter, ammonium salts, nitrates, water of crystallization, etc. Determination of OO2 in Soil Air. — The analysis of soil air is conducted upon the same principles as that of ordinary air, but the method employed is necessarily different so far as the obtaining and handling of the sample are concerned. The reagents are the same as required in the analysis of atmospheric air ; the apparatus, however, is quite different. It consists of a number of sections of water-pipe with screw joints, one having a pointed foot, above which are a number of perforations within a limited area ; an absorption tube, in which the barium hydrate solution is held and through which the air is drawn, and an aspirator. (See Fig. 23.) The section with the pointed end is driven into the soil, and the pipe is lengthened by the addition of the other sections, so that any desired depth may be reached, and thus the air of any stratum may be withdrawn. The upper extremity is connected by a rubber tube with the inlet tube of the absorption apparatus, which latter may be a plain glass tube about an inch in diameter with a bend of about 130 degrees near one end. Better, however, is the apparatus shown in the illustration. Here the short leg of the bent tube is a large bulb, and the long leg is a series of small bulbs, the communications between which are of small diame- ter. In either case the inlet tube passes through a tightly fitting rubber stopper and extends to a point just beyond the bend. The other end of this apparatus is connected by means of a rubber tube with the inlet of the aspirator. Any form of aspirator may be used, but preferably one of a capacity of about twenty liters. A measured amoimt of the dilute solution of barium hydrate, sufficient to occupy the greater part of the long leg, is introduced into the absorption appa- ratus, and the connections throughout are tested to prove the absence 314 THE SOIL. of leaks. When the outlet cock of the aspirator is opened, the escape of the contained water creates a partial vacunm, which is relieved by suction of air from the soil and through the whole apparatus. As the air emerges from the inlet tube of the absorption apparatus, it passes upward in the form of bubbles through the reagent, to which it gives up its content of CO2. The reason for preferring the bulbed tube is that each bubble of air in its passage from one bulb to the next above is necessarily brought into more intimate and prolonged contact with the reagent than is the case when the plain bent tube is employed, for Fig. 23. ^ % \ - 00 P'S'^ ■ ■" -i- , *^ Apparatus for determination of CO3 in soil air. here the air bubbles pass quickly along the upper inner surface of the tube, and are not so exposed to the reagent as to lose all the contained CO.,. For this reason, it is necessaiy to draw the air through a second, and, perhaps, a series of such tubes, but one bulbed tube as pictured above is sufficient. The water from the aspirator is measured carefully, and its amount indicates the volume of air that has been sucked uj) out of the soil to take its place. AVhen the desired amount has been acted u})on, the stopcock of the aspirator is closed, and the reagent in the absorption BACTERIOLOGICAL EXAMINATION OF SOIL. 315 tube is transferred quickly to a glass-stoppered bottle of suitable size. From this point, the determination is the same as described in the chapter on Air. Bacteriological Examination of Soil. The bacteriological examination of the soil requires necessarily an intimate acquaintance with bacteriological technique, a subject beyond the scope of this work. It may be stated briefly that many of the organisms that inhabit the soil may be isolated by adding small por- tions of the sifted sample to liquefied gelatin and then plating, or by sprinkling over the surface of a nutrient medium, or by shaking with distilled water and transferring thence to the proper media. The many anaerobic forms requu'e, of course, the special treatment of their class, and some of them may be grown on ordinary culture media ; but many of the saprophytes, notably the nitrifying organisms, cannot be isolated by the ordinary methods. For the details involved in the separation and identification of the numerous varieties of soil organisms, the reader is referred to the standard works on bacteriology. CHAPTER IV. WATER. Absolutely pure water, that is, the substance composed wholly of hydrogen and oxygen, and represented by the symbol 11,0, is never found in nature, and is never seen, except in small amounts as a labora- tory curiosity. In the broad sense, however, the word jiure as applied to water conveys the idea of freedom from harmful ingredients and of wholesomeness and suitability for drinking and for the preparation of food. In nature, all water contains more or less of gaseous and solid substances in solution and suspension, and so long as these are not present in such amounts as to aifect the quality injuriously, and so long as they are not intrinsically dangerous to health, the adjective is com- monly held to be appropriate. But in the sense that purity involves the limitation of the amount of contained substances of a harmless nature, it becomes a diificult question where to draw the line where water ceases to be pure, and what term to apply as an antonym. In the sense that it involves complete absence of matters intrinsically dangerous, the line can be sharply drawn, and water which fails to satisfy the requirements of the term may be designated indifferently as impure, ])olluted, or contaminated. In the classic reports of the State Board of Health of Massachusetts on public water supplies, waters are classed as " normal " or " polluted '* according as they are or are not free from direct or indirect pollution by the waste products of human life and industry. Under this classi- fication it follows, naturally, that normal waters must vary veiy Avidely in appearance, composition, and general character, and that a normal water is not necessarily suitable for drinking, although incapable of causing specific disease. The nature and amount of the dissolved matters cannot but have considerable influence in modifying the prop- erties and effects of a water. Waters may be classified according to source as follows : 1. Rain and snow. 2. Surface-water (rivers, ponds, basins, etc.). 3. Ground-water (also known as subsoil-water). 4. Artesian or deep well-water. RAIN. Rain is the original source of all natural waters of whatever class. It results from condensation of the aqueous vapor of the atmosphere, and in its descent to the earth it takes up gaseous and suspended mat- ters from the atmosphere, which to that extent becomes thereby puri- 316 SURFACE-WATERS. 317 fied. In the open country, after the air has been washed for a while, the collected rain is very clean, and is, in fact, the purest form of nat- ural water. If its fall is accompanied by wind from dusty localities, it cannot be obtained in so clean a condition within so short a time, on account of the greater amount of suspended matters to be washed down. Near the sea, it contains more or less salt ; and in cities and large towns, it may have a slightly acid reaction. In its passage downward through the atmosphere, rain absorbs con- siderable air, or, more properly, constituents of air ; that is, oxygen, nitro- gen, carbon dioxide, and ammonia compounds. Since each gas has its own coefficient of solubility in water, and as air is a mixture and not a chemical union of gases, it follows that water will absorb the con- stituents of air separately and according to their respective solubilities. So it happens that the absorbed air has a very different composition from that of atmospheric air, being much richer in oxygen and poorer in nitrogen, its oxygen content being 35 instead of 21 per cent. On reaching the earth, some of the rain is evaporated, some sinks into the soil, and some runs over the surface to streams or other bodies of water. The amount that sinks into the soil depends upon the permeability of the latter to water. Thus, a sandy or gravelly soil will take up more of the rainfall than a close-grained clay. The amount which is returned to the atmosphere by evaporation is surprisingly large. It has been reckoned by Dalton that in the whole of England and Wales, about 50 per cent, of the total annual rainfall is lost by evaporation. In the watershed of the Rhine, the loss is reckoned at 50 per cent. ; in that of the Rhone, at 42 ; of the Seine, at 67, and of the Garonne, at 35. SURFACE-WATERS. Surface-waters are collections of water running along or stored upon the earth's surface in contact with the atmosphere. Under this head are included rivers and smaller streams, ponds, lakes, and impounding basins. They vary according to the different characters of the areas which they have drained or traversed, or in which they are stored. Thus, a water that has flowed over a rocky soil is more likely to be free from organic impurity than one that has flowed over loamy soil or has stood in swamps ; and one that has flowed through sandstone bot- toms is more likely to contain mineral impurities than one that has flowed over the virgin soil of a forest. Surface-water means something more than the rain of the district plus the impurities of whatever character, organic and mineral, which it has collected. Rivers and lakes, for example, are made up of rain that has run over the surface of the ground, dissolving in its course small amounts of easily soluble matters, and of water that has come up from the soil below through springs, or that has trickled in from the upper layers of the soil ; and these latter contribute matters which may lae of very widely different character from those obtainable along the 318 WATEB. surface, according to the geological character of the soil strata that have been acted upon. A river may take its origin in a spring, and consist for some time of ground-water alone, but usually it is not long before it receives acces- sions of surface-water and soon acquires the characteristics of the latter. Again, some lakes and ponds are fed almost wholly by springs at their bottoms and sides ; but even so, their waters soon change in character and acquire the various forms of aquatic life. Surface-waters may contain much or little or no organic matter, according to circumstances. They may be colored or colorless ; they may be rich or poor in mineral substances. Those which come largely from the ground will naturally possess largely the characteristics of ground-water, and those free from accessions from this source will approximate more nearly the character of rain. The quality of surface- waters is influenced by the seasons, by drought and rainfall, by vege- tation, by rate of movement, and by other conditions. GROUND-WATERS. Ground-water is that which penetrates the soil, sinks to various depths, according to the nature of the soil, and accumulates on some more or less impervious stratum. It is not exposed to light and the atmos])here, like surface-water. It varies widely in character acoord- iuff to the nature of the soil over which it has once flowed and through which it has percolated. It enters with more or less air and CO^ in solution, and comes in contact with the soil air in the interstices, which is much richer than atmospheric air in this gas. With the assistance of the CO^, which it has brought, and that which it farther acquires in the interstices, it dissolves various mineral constituents of the soil. That which penetrates very deeply has its solvent power increased by increased temperature and pressure. As it enters the soil, it brings with it whatever organic matters it may have dissolved out of the surface layers, and in its descent it may lose them entirely through the action of the saprophytic bacteria of the soil, or it may acquire still more if the soil be polluted and so permeable as to permit rapid passage downward. It passes slowly or rapidly through the interstices until it reaches an impermeable stratum, over which it accumulates, filling the interstices completely. The soil at this point is said to be saturated, and the upper limit of saturation is knoAvn as the ground-water level, or water table. Between this and the surface, the water is in contact with the air of the interstices, and is known as capillary moisture. The water table is by no means necessarily hori- zontal, but follows in a general way the contour of the surface of the soil, and often it is much more irregular, and, by reason of local geo- logical conditions, even quite different from what the surface formation Mould indicate. Thus, at one point in a level stretch of country, the table may be quite near the surface, and at another, a short distance GROUND- WATER. 319 away, it may be situated much more deeply, owing to abrupt changes of level of the impermeable stratum. Irregularity of the surface of the water table is due largely also to the rainfall, which, coming at frequent intervals, falls upon surfaces of diifering permeability, so that while one part is still draining its water downward, another has completed the process and is ready for more. When drought occurs, however, the level becomes more and more uni- form until it may become quite horizontal. AVith return of rainfall, the level rises, and irregularity of the surface of the water table is again produced. The level at any point is influenced also by the amount of water withdrawn from the soil by the demands made upon wells. When the amount of percolation is exceeded by the amount of withdrawal, the level falls ; when the conditions are reversed, the level rises. The water table in its irregular course touches the surface of the ground here and there, and gives rise to springs which may flow the Fig. 24. Outcropping of water table. year round regardless of drought, or may dry up completely with fall of the level. Similarly, all permanent ponds are outcroppings of the water table, and the beds of rivers as well, but the level of the table in the near vicinity is almost invariably higher than the surface of these bodies. Sometimes, however, the water level is so near the sur- face that, without emerging in the form of springs, it extends in a broad sheet just at or below it and causes marshy conditions. In Fig. 24 the manner in which the water table crops out in springs and feeds lakes and other bodies of water is sho'^m. By some, the water table is spoken of as an underground river, a term which is very misleading, in that it suggests a body of water rather than a condition of saturation of the soil. There are, to be sure, in some localities, especially in limestone districts, bodies of water flow- ing between impermeable strata, and instances are known of disap- pearance of streams into fissures of rocks and emergence at a distance elsewhere, but these streams are not a part of the water table as gen- erally understood and may not properly be classed as ground-water. In most cases, and except where the water Hes in deep depressions 320 WATER or pockets with no side outlets, the ground-water is in constant lateral motion in the direction of the outfall, and this is commonly the nearest large body of water, either a lake, or a river, or the sea. In its onward course over an irregular impervious stratum, the movement is at times inclined upward and at times dcwTiward, but ever in the same general direction laterally. The rate of movement is determined by a number of influences, among which the most effective are the degree of permeability, the inclination, and the barometric pressure. The degree of permeabilitv, dependent upon the coarseness of the soil particles, is of very great importance, the more rapid flow occurring through the soils of coarser texture. The inclination, or, in other words, the influence of contour in promoting or preventing the assistance of gravity, has a very decided effect. The barometric pressure affects the rate of movement through its effects on the air in the interstices above the water level. While this air is itself in constant movement, it cannot move quickly because of the great amount of friction created. Lessened pressure above the ground causes the soil air to expand, and as this occurs, the tendency is along the lines of least resistance, namely, upward and, under cer- tain conditions, laterally, so that the water in the interstices is assisted in its flow. But the influence of diminished barometric pressure is felt almost at once at the outfalls, because of lessened back pressure on the water. This influence may be measured by noting the fluctuations in the water levels in wells which rise as the barometer falls, and fall as it rises. Thus, resistance is removed at the outfall, and coinci- dently the water is being pushed along by the expansive force of the air in the interstices. AVith increased barometric pressure, these condi- tions are reversed and the flow becomes less rapid. The rate of movement being so dependent upon local conditions, it follows that it varies widely in different soils. In some places it is so slow as to be almost unmeasurable ; in others it is extremely rapid ; and even within a restricted area, it may be exceedingly variable at different points. At Budapest, for example, Fodor ' determined the rate of movement at five different points to be 95, 125, 199, 209, and 210 feet daily. The average of these figures, 167.6, represents unusually rapid flow. At Munich, the daily rate of flow toward the Isar has been calculated by Pettenkofer as a trifle more than 15 feet. Physical and Chemical Characteristics of Water. At the standard barometric pressure, 760 mm. or 29.922 inches, water boils at 100° C. or 212° F. With lower pressures, it boils at correspondingly lower temperatures ; on very high land, for example, it boils at such low temperatures that meat and vegetables cannot be thoroughly cooked in it. Evaporation occurs at all temperatures, even below the freezing-point. ' Boden und Wasser. Brunswick, 1882. PHYSICAL AND CHEMICAL CHARACTERISTICS OF WATER. 321 Water has its maximum density at 4° C.^ above and below whicb point it expands. At 0° C. it freezes, and in doing so, it expands to the extent of about 9 per cent, of its volume, and thus acquires a specific gravity less than that of unfrozen water, in which, therefore, it floats. As the surface freezes, it gives out heat to the layer immedi- ately beneath and thereby causes a retardation of the process. As this layer becomes cooled, the ice formation continues, and thus the growth in thickness of the ice cover proceeds downward. Its specific heat is high, and is taken as the standard of comparison. As a conductor of heat it stands very low. Water is the most universal solvent known, there being but few sub- stances which are not acted upon by it to some extent. It takes up all known gases, and its solvent power for them is greater according as the temperature is depressed and the pressure increased. In the case of substances other than gases, with few exceptions its solvent power is increased with increased temperature. Appearance. — Pure water is clear, and, in proportion as it contains dissolved air and carbon dioxide, is bright and sparkling. Brilliancy of appearance is, however, by no means conclusive evidence of purity, some extensively contaminated waters showing remarkable brightness. Turbidity of water is due to organic and mineral matters in suspen- sion ; the organic matters may be ordinary dead vegetable and animal substances or microscopic living plants and animals. Some public supplies derived from rivers are distinctly muddy in appearance. The slight degrees of turbidity designated as milkiness and opalescence are due commonly to very minute clay particles, which may remain in suspension for a long time, even when the vessel con- taining the water is allowed to stand undisturbed. Sewage matters also may give these same appearances. Turbidity due to clay may be removed readily by the addition of various substances, as lime, alum, and sulphuric acid, which cause the particles to agglutinate and settle out. Water which is apparently clear when viewed in an ordinary glass vessel may be seen to have decided turbiditv^ when viewed through a depth of a foot or two against a pure white surface. Some ground-waters which are cpiite clear when drawn may acquire a turbid appearance on standing, due to the presence of compounds of iron which undergo changes in composition and become precipitated. In such cases, the turbidity is accompanied by the development of color, which, however, disappears on the completion of the process of oxidation of the iron compounds and their separation by sedimentation. Color. — Water may have color or not, according to circumstances. Surface-waters may derive it from contact with grasses, leaves, woody matters in general, and peat, the degree of color being dependent upon the length of time of contact and upon the character of the substances. Different kinds of leaves, for example, impart different shades and kinds of color, but not always to the extent that their appearance would indicate. The dark-colored dried leaves of the oak, for instance, might be expected to yield a much darker infusion than the much 21 322 WATER. lighter colored leaves of the maple, but such is not the ease, as may be proved readily by experiment ; and those of the butternut give a color that is surprisingly light in comparison. Long contact with swamp vegetation causes a deep reddish-brown color, Mhich is often very stable on long keeping. Xot all surface-waters, however, are exposed to color-imparting substances, and waters of this class may be free from color. Ground-waters of good quality are ordinarily colorless or appear to have, when viewed through considerable depths against a white surface, a faint bluish or greenish-blue tinge. Sometimes they contain iron and orgtuiic matter in combination, and have in consequence a brownish tint, which, by reason of very slow oxidation, may jK-rsist for a long time. Color derived otherwise than from contact with vegetable matter is accompanied usually by more or less turbidity. Absence of color is not a sign of purity, for jiolluted waters may be quite free from it ; nor is its presence an indication of unfitness for domestic use. Reaction. — The dissolved carbon dioxide in water tends to give it a slightly acid reaction, but most potable waters are verj^ faintly alkaline to delicate indicators, owing to minute amounts of alkaline carbonates. Rain-waters, especially in the vicinity of cities and large towns, are generally slightly acid on account of impurities of the atmosphere, arising from combustion. Peaty waters also are slightly acid on account of organic acids produced by the action of the peculiar bacteria existing in peat. River-waters in mining districts often contain con- siderable amounts of free mineral acids. Odor. — Pure water has no odor, but good surface-waters containing coloring matters have more or less odor, which is especially marked on heating. It is generally suggestive of vegetable matter, and may be characterized variously as grassy, peaty, etc., according to the impres- sion ]>roduced. Such odors may persist even on long boiling, while those due to dissolved gases will disappear quickly on heating. ^lauy othei'M'ise good surface-waters are jiarticularly prone to the develop- ment of disagreeable odors attributable to minute living organisms. The subject has been studied very extensively by Mr. Gary X. Calkins,^ Avho states that odors in drinking-waters " may be produced by the putrefactive decomposition of the body plasm through the agency of Ixicteria, and by the excretion of certain products of groAvth, or by the liberation of products by the physical disintegration of the body or breaking down of the enclosing cell walls. These three causes give rises to three classes of odors, as follows : (1) odors of chemical or putrefactive decomposition, (2) (^lors of growth, and (3) odors of pliysical disintegration." The group of plants popularly known as '^ blue-green algse " (Schizophi/cecv) is a very common cause of the well-known " ]>ig-pen " and " grassy " odors so frequently observed in shallow, stagnant, and relatively warm waters. Certain of the Dkdomacea' frequently cause serious trouble by imjxirting aromatic (geranium) and fishy odors ' Kopoit of the State Boiird of Health of Massachusetts for 1892, p. 355. PHYSICAL AXD CHEMICAL CHARACTERISTICS OF WATER. 323 and disagreeable taste. Of these, the most prominent is Asterionella formosa, found very commonly in large ponds and reservoirs of surface- water, and growing with especial luxuriance in open reservoirs of ground- water. According to Whipple and Jackson,^ 3000 asterionella per cc. of water may, under favorable conditions, impart an odor easily recog- nized by the consumer. Several species of Uroglena, commonly, but according to G. T. Moore,^ perhaps incorrectly, classed with the Infu- soria, cause much trouble by the liberation, during disintegration, of oil globules which impart fishy, oily odors and tastes. These oil globules are yielded by many other varieties of water organisms. AVhile sewage matters impart mouldy or musty odors to water, it should not be inferred that these odors are of themselves indicative of sewage pollution, for good surface-waters sometimes acqmre them on standing. Sometimes it will be noticed that water on long boiling not only con- tinues to evolve a vegetable odor, but gives it off in greater intensity. This is true particularly of waters rich in algse. If they are first filt- ered, the odor will not be given off on boiling. But other waters may continue to evolve odors even after filtration. Peaty waters, for in- stance, often persist in yielding odor on long boiling, and this is not affected in any way by filtration. Waters containing products of physi- cal disintegration and various other substances also are not influenced by filtration. Odors which disappear on boiling may develop again after a time if their cause is not removed ; if, however, the matters from which they are derived are no longer present, the odor will not return. Some most troublesome odors are known to be the results of decay. The public supply of Boston was, in 1878, seriously affected in this way, and gave off an odor which was likened to that of cucumbers. This was investigated by Professor Ira Pemsen, who found the cause to be decomposition of a fresh-water sponge. Water sometimes contains sulphuretted hydrogen from reduction of sulphates by bacterial action, and sometimes mixtures of products of organic decomposition which suggest that gas. Very marked and most offensive odors are due often to the presence of dead animals, such as toads and mice in wells, and, when they arise, the remedy is ob- vious. Some wells become stagnant at the bottom, and if organic matter is present, it may cause foul odor, suggestive of dead animals, by putrefaction in the absence of a sufficient supply of dissolved oxygen. Stagnation may be prevented by connecting the pump nearer the bottom, or by filling up the unnecessary space with clean gravel and sand. Odors in water are not necessarilv indicative of dano-er to health, but distmctly unpleasant ones are c[uite sufficient as a disqualification, on account of the repugnance which their use for drinking and other domestic purposes would cause. On the other hand, as in the case of ^ Journal of the Xew England Water-"\Vorks Association, September, 1899. ^ American Journal of Phai-macy, January, 1900. 324 WATER. color, absence is not indicative of purity, for dangerous waters may be inodorous. Taste. — Pure water has no distinct taste, and, whatever the impres- sion made, it is due to dissolved gases. That this is so, is most evident when one compares the taste of a well-aerated water, before and after heating to the boiling-point with subsequent cooling. Saline constituents impart no distinct taste unless they are present in quite large amounts, as in waters of a high degree of permanent hardness. The only substance which imparts taste when it is present in very small quantities is iron. Dissolved organic matters cause no taste, un- less present in considerable amount and, as a rule, accompanied by odor. AVater containing very little coloring matter is often said to taste distinctly, but it should be remarked that the senses of taste and smell are often mfluenced unconsciously by the sense of sight, and colored water supposed to have both odor and taste may, if drunk in the dark, give no impression of either. Badly tasting water, whether dangerous or not, is objectionable on the same grounds as mentioned under odor. Not only is absence of bad taste no evidence of purity, but it is well known that waters con- taining the products of oxidation of sewage are often remarkable for unusual palatability. Substances Found Normally in Water. These include : 1. Gases in solution. 2. Organic matters in solution and in suspension. 3. Mineral matters in solution and in suspension. 1. Gases. — First in importance is air. Strictly speaking, water con- tains no air as such, but only the constituents of air, for the oxygen and nitrogen, dissolved by water, are not present in the same propor- tion in Avhich they exist in the atmosphere. In salt water, the varia- tions in their proportions are less wide. We shall, however, consider the two gases as air. The dissolved oxygen is the important element. One hundred volumes of water at 15° C. will dissolve nearly 3 vol- umes of oxygen (2.99), and at 20°, 2.80 volumes, and it is not alto- gether removed by boiling. The amount of oxygen in solution is fairly constant in waters of uniform composition freely exposed to the atmosphere, but when they receive additions of sewage and other oxidizable matters they begin to lose it. Eiver-waters may thus show notable differences in the amount of dissolved oxygen present in samples taken above, within, and beloAV towns situated on their banks. The Thames and the Seine, for in- stance, show this in a remarkable degree. The progressive diminution is due to the constant access of organic matter, which undergoes oxida- tion at the expense of the dissolved oxygen. AVhen a river-water is deprived of its dissolved oxygen in this manner, or by reason of chemical changes due to the inflow of sewage from manufacturing SUBSTANCES FOUND NORMALLY IN WATER. 325 establishments, containing compounds — ferrous, for instance — ^having a strong affinity for oxygen, fish life cannot be supported. Absence of fish in polluted streams is due much more to diminution of dissolved oxygen than to the poisonous effects of organic sewage. Aeration of water is influenced very largely by the dust which falls into it, for each particle carries with it more or less adherent air, as may readily be seen when one drops small particles into water and observes their descent. Aeration of water proceeds to great depths, as is shown by chemical analysis of samples of water obtained by de^p sounding, and also by the fact that great numbers of organisms which require oxygen for their respiration are found far beneath the surface ; but water at 40 and 50 feet below the surface may contain no oxygen. Water from deep wells is very commonly free from dissolved oxygen, because of abstraction by compounds of iron or manganese, organic matters, and other substances. The presence of considerable dissolved oxygen in water leads to beneficial changes in the organic matter present. Diminished oxygen permits the development of low forms of vegetable life, which fre- quently give rise to unpleasant tastes and odors. Their growth is inhibited by a large degree of aeration, and their disagreeable effects are thereby prevented. Carbon Dioxide. — The carbon dioxide contained in water is derived largely from the atmosphere, and in great part from the soil, where it is present in abundance. Its amount in any water depends upon a number of circumstances : upon the amount carried in by rain and dust, the character of the soil, and the extent of oxidation of organic matter occurring in the interstices. It is greatest in amount at great depths, and it may constitute almost the entire content of dissolved gases. It has been calculated that the ocean contains about ten times as much as the entire atmosphere. 2. Organic Matter. — The organic matters in water are of both ani- mal and vegetable origin, and consist of organisms, products of organic life, and results of disintegration and decomposition. The animal matters include dead and living organisms and dissolved and sus- pended products of animal life and decay, such as albuminous sub- stances, urea, and tissues. In the tropics and subtropics, ova and young of various parasites are common. Ordinarily our interest in organic matter from animal sources is confined to the products of human life as represented by sewage, which may contain the exciting causes of specific diseases (see Bacteria in Water, page 332). Vege- table organic matter exists as living and dead organisms and tissues in suspension, soluble and suspended substances given off during life, and soluble matters extracted by the water after death. The vegetable organisms are represented by very numerous species of microscopic plants, which act beneficially by absorbing the products of organic decomposition for their growth, iDut which may, on the other hand, under fiivorable conditions, become the source of much trouble by over-abundant growth, disintegration, and decay. They may prop- 326 WATER. erly be regarded as normal constituents of surface-waters, for they are always present in such, and, moreover, they develop quickly in stored ground-water exposed to light and air. When they die, most species appear to decay rather slowly, and the products of their decomposition are absorbed by new growths ; but when ])resent in great abundance, the progress of decay may exceed that of growth, and then their products may accunuilate and cause foulness. There is one form of microscopic organisms, belonging to the class of fungi, which merits special mention : Crenothrix Kithniana. This is a filamentous plant with cells no larger than the ordinary bacteria. It grows chiefly in ground-waters which contain organic matter aud iron, the latter of which ingredients it fixes in the form of ferric oxide in the gelatinous sheath of its filaments, which thereby become yellow, yellow brown, or brown in color. It causes great annoyance by the rapidity with which it grows in water-pipes, the lumen of which is not infrequently completely occluded. This may occur more readily where the surface presents roughness and imperfections, to which the growths may attach themselves. When the filaments are broken oif and become disseminated through the water, the latter is rendered unfit for laundry use on account of the iron-rust. Sometimes, it gives rise to disagree- able odors and an inky taste. It may be very troublesome within the tubes of driven wells, or in the reservoirs, as well as in the distributing pipes. Sometimes, it may be seen in large aggregated masses floating about on the surface of stored water. By its extensive growth in pipes, it may seriously affect a whole public supjily. The presence of living forms, either vegetable or animal, indicates that the Avater contains at least whatever food materials are necessary for their existence, but not necessarily that these are in excess, Algse, for instance, require mineralized nitrogenous matter (nitrates), aud other substances ; fungi suggest th(> jiresence of carbohydrates, i)ro- teids, and mineral substances common to domestic sewage ; infusoria suggest organic decomposition. Dissolved vegetable matters ordinarily amount to but little in weight. Even in some very brown waters, whose appearance would suggest large amounts, they may be present to the extent of not more than 1 or 2 parts in 100,000. The organic matters, both animal and vegetable, which are of inter- est to the sanitarian, consist chiefly of carbon, hydrogen, oxygen, and nitrogen, with, in many cases, small amounts of ])hosphorus and sul- phur. In the process of decomposition, which owes its inception, progress, and completion to bacterial activity, the carbon is cond)ined with oxygen to ft)rm carbon dioxide, and the hydrogen unites in ])art with nitrogen to form ammonia, the jjresence of w liich in water in- dicates that the process of decomposition is under way. In its turn, as will be shown later, the ammonia is converted eventually to nitric acid, which unites with bases to form nitrates. Ammonia. — From the standpoint of sanitary significance, ammonia in water is of prime importance. Only under very unusual conditions does it exist in the fi)rm of hvdratc, but usually as chloride or car- SUBSTANCES FOUND NORMALLY IN WATER. 327 bonate. We speak of it commonly as free ammonia, for, on boiling the water, these salts are decomposed and the ammonia is expelled in the steam. Among the direct sources of ammonia in water is rain, which brings it down out of the atmosphere in varying amounts according to location. Eain always contains it, but more is present in that of thickly populated districts than in the open country. In one instance, reported by Drown,^ it was found to the large extent of 0.0564 in 100,000. Its presence, however, in surface- and ground- waters is due for the most part to decomposition of nitrogenous organic matter. It is not abundant in ordinary unpolluted waters, but is present often to a very considerable extent in that of deep driven wells. Here its origin is not always clear ; in some cases it is supposed to be referable to coal deposits, in others to reduction of accumulated nitrates. Under ordinary conditions in surface-waters, ammonia, after conver- sion to nitrates, is absorbed very quickly by growing vegetation, and the more active the conversion and the growth, the greater the appro- priation. For this reason, water from the same source will often show less on analysis in summer than in winter. But activity of vegetation is not responsible alone for this difference in amount, for in the case of large bodies of water, as lakes and ponds, the rate of movement of the water has great influence. During the warmer months, when the upper layers are warmer and consequently lighter than the lower, the latter become necessarily stagnant and stratified. The ammonia which accumulates in these lower strata does not, therefore, come to the sur- face until cold weather approaches. Then the upper layers become more dense and tend toward the bottom, causing a displacement of the lower layers toward the surface and general uniform mixing of the entire volume of water. Another element in the stirring up of the water of ponds and lakes is the action of wind, which, however, does not extend beyond twenty feet. Still another influence to be con- sidered is that of springs at the bottom and sides, which tend to keep the water in motion. In the case of flowing rivers, the water is of comparatively uniform composition at all depths. Ammonia is very characteristic of sewage pollution, the oxidation of which yields it in abundance under conditions which do not permit it to be rapidly oxidized to nitric acid. Ammonia as it occurs in drinking-water is of itself incapable of producing harmful effects. Its amount, however, is of greater or lesser significance according to circumstances : that from clean and properly stored rain-water is of far less significance than that from other waters. In the one, it may be considerable in amount and mean but little ; in others, it is usually evidence of decomposition of organic matter. Its amount in good water is not large, and on account of oxidation and absorption by vegetable growth it does not accumulate. And even in sewage-polluted waters, when vegetation is active, oxidation and ^ Massachusetts State Board of Health : Keport on Water Supply and Sewerage. Boston, 1890. Part 1, p. 562. 328 WATER. absorption may so diminish its amount that, taken alone, it might lead to false conclusions as to the character of the water. Albuminoid Ammonia. — The so-called albuminoid ammonia is am- monia which is produced in the process of analysis of water by the action of alkaline permanganate of potassium on nitrogenous organic matter hitherto undecomposed. The result of the action is a splitting up of the organic matter and the conversion of the nitrogen to ammonia, "which, as is the case with " free " ammonia, passes out of the water in the steam. This matter may be of either animal or vegetalde origin, and its character is of far greater importance than the amount of the yield. Thus, a water grossly polluted by sewage may yield less than another quite free from such contamination, but rich in dissolved vegetable matter of no great sanitary importance. Animal organic matter is decomposed much more rapidly than veg- etable matter, some kinds of which are remarkably permanent, such, for instance, as the substances which impart the brown color to the waters of swamps. Animal matter is richer in nitrogen than vege- table matter, and consequently a stated amount of all)uminoid ammonia represents decomposition of a larger amount of the latter than of the former. In other words, a small amount of animal matter will yield as much albuminoid ammonia as a large amount of vegetable matter. Inasmuch as animal matters are of far greater significance than vege- table matters, it must be clear that the amount of albuminoid ammonia is of less importance than its origin. And since, in the analysis of water, the ammonias themselves give no indication of their origin, their significance can be measured on]\- with the aid of estimations of other substances ; and often, also, a knowledge of tlie source of the water and its surroundings will be required. Nitrites and Nitrates. — The ammonia formed in the first stage of decom])ositi()n and that washed out of the air by rain are oxidized eventually to nitrates under the influence of the so-called nitrifying bac- teria, and this stage marks the comj)letion of the process. The nitric acid formed, coming in contact with earthy and alkaline carbonates, attacks them and unites with the bases to form nitrates and, in so doing, liberates carbon dioxide. The nitrifying process occurs not alone in the body of the water itself, but to a much greater extent in the interstices of the' soil, so that a water rich in all manner of organic substances undergoes, under favorable conditions, this purifying process in the fullest degree when it enters the soil at the surface and percolates slowly downward. Before the stage of complete nitrification is reached, tliere is an intermediate stage, that of nitrous acid and nitrites, but it is probable that the time during which a given amount of nitrogen on its way to mineralization remains in the nitrous form is extremely short ; in fiict, the step from anuiionia to nitric acid is practically instantaneous. Nitrates are seldom absent in either surface- or ground-waters, and may be present, especially in the latter, in quite large amount (as mucli as (3 or 7 or more parts in 100,000) ; while, on the other hand, nitrites are SUBSTAXCES FOUyD XOEMALLY IX WATER. 329 not ordinarily present in unpollnted waters, and as little as y-wtq P^^^ in 100,000 of any water is looked upon as ''high.'' It is a fact that nitrates are reduced xery readily to nitrites, and farther back to ammonia, and even to nitrogen gas itself, by a variety of organisms which act in the absence of oxygen. These are known as the denitrifying bacteria, and while these species are doubtless very numerous, only a Irmited nimiber have been isolated and identified. Their action is inhibited by oxygen, as has been proved by Stutzer and Maid,' who found that the process of denitrification ceases in cultures through which a stream of oxygen is passed. This was confirmed by Weissenberg, who observed, further, that when the bacteria Avere culti- vated in small volumes of nitrate bouillon in flasks of such shape that the surface of the liquid was very great in comparison to its depth, and exposed to the air, they did not act. These bacteria are common in sewage in which the conditions for their growth and activity — absence of dissolved oxygen, for instance — are present. Grimbert^ has shown that B. ti/phosus and B. coli com- munis reduce nitrates and amido principles in culture media. The production of gas appears to be a result of the secondary reaction on the amido compounds by the nitrous acid formed through bacterial action. Small amoimts of nitrites in water may be derived from the air by absorption or by the cleansing action of rain, and may be due to con- tact of metallic surfaces, brickwork, and new masonry with the nitrates in solution ; but they are almost never present in what are called large amounts (one part in a hundred million) except as an indication of sewage pollution. The disproportion between the amounts of nitrites and nitrates in water may also, perhaps, be explained as follows : The nitrates are the final stage of complete oxidation ; they do not go on to a higher form, but, being permanent in character, accumulate in the water, unless with- drawn bv vegetable life or reduced. The nitrites cannot accumulate as such, but are converted to the higher form. Thus, the lower form is constantly passing into the higher, and is stored as such. Nitrates vary considerably in amount, owing to various causes. They are almost always present in both siu'face- and ground-waters, unless there is some process at work causing a reduction to nitrites. In impolluted surface-waters they are usually low in amount, but such waters generally contain more nitrogen in this form than as ammonia. They do not accumulate greatly in such waters during the warmer months, for they are absorbed largely by growing vegetation. Hence they are more abundant in winter. In the warmer mouths they may be absorbed almost wholly by growing algte. Ground-waters contain little or much, according to circumstances ; in virgin and thinly settled districts the amount is small ; in others, it is usually fairly high. In the former, it is mainly from the ammonia 1 Centralblatt fiir Bakteriologie, Abth. II., Bd. 2, 1896, p. 473. ^ Annales de I'lnstitut Pasteui-, Jan., 1899. 330 WATER. of the rain and that formed in the decay of the organic matters nat- urally in the .soil ; in the latter, it is due largely and mainly to the ammonia of domestic sewage. Ground-waters rich in nitrates, when exposed to light and air, generally become more or less rich in vegetable growth, and poorer in nitrates. Like ammonia, nitrates in water are not of themselves in any way harmful in the amounts found. They simply represent what was once organic nitrogen, but now completely mineralized. Xor is their pres- ence any indication of the nature of the original organic matter, whether animal or vegetable, and this can be inferred only when other constit- uents are considered. When present in considerable or very high amounts, they indicate a corresponding degree of paxt pollution, per- haps nearby existing pollution, and the possibility of future danger from its recurrence. Therefore, high nitrates should sometimes be looked upon with suspicion. And, furthermore, it should be borne in mind that the evidence of extensive mineralization does not preclude the existence of present processes and the presence of active pathogenic micro-organisms, for organic matter may be oxidized rapidly in the presence of living patho- genic germs. Sometimes, veiy large amounts of nitrates are found in the waters of very deep wells, so large that they cannot be explained by the supposition of oxidized sewage. In these cases the cause is surmised to be fossil remains or natural nitrate deposits. The presence of nitrites in water is of far greater importance than that of nitrates. It means that fermentative changes are in progress, and that oxidation is not being completed. When this condition obtains, nitrites may be very persistent. Sometimes, they mean a reduction of the nitrates, which takes place mainly under the influence of deuitrifying organisms, quite likely to be present in large numbers in decomposing org-anic matter. Sometimes, neither nitrates nor nitrites are present in sewage-polluted water ; in such cases, either they have not been formed or they have been completely reduced. When nitrites are present at the expense of the nitrates Ijy the action of metallic surfaces, lead and iron, for example, the metals themselves are present in at least detectable traces. 3. Mineral Matters. — Chlorine as common salt is a normal constit- uent of all wat«a>. luiiu-water take- it n\) from the air in small traces, particularly near the sea coast. In the specimen of rain referred to on page 327 as rich in ammonia, the chlorine content was 0.13 per 100,000, which is much in excess of that found in nuiny inland waters. The amount of chloriue normally jiresent in the water of a district dejiends on location and other couditions. It is influenced very greatly by proximity to the sea, the air above which contains necessarily more than tliat at a distance inland. It varies in amount in the same water with differences in the amount of rainfall and evaporation, and in the direction of the wind. Chloriue increases du'ectly with the population, and its amount is SUBSTANCES FOUND NORMALLY I^ WATER. 331 influenced veiy greatly by a proper system of sewerage which carries the sewage matter, rich in common salt, beyond the limits of the drain- age area. When its amount rises aboye the normal of a locality, it is indicatiye of sewage, though not necessarily of recent pollution. As we haye seen, the organic matters become mineralized, and no longer exist in their original form ; but no such change occurs in the chlorides, which remain fixed and unchanged, and they may be the only eyidence remaining. Thus a water polluted by sewage may haye its organic nitrogen conyerted to nitrates, aud these in turn may be absorbed by yegetable growth ; it may be clear, colorless, odorless, and palatable, free from pathogenic bacteria, and in eyery way suitable for drinking, but, neyertheless, the chlorine remains as a witness that pollution has occurred in the past. According to Professor Drown, in a general way 4 families, or 20 persons, per square mile will add on an ayerage 0.01 part of chlorine per 100,000 to the water of a district in seasons of ayerage flow, and more in time of drought. Other Mineral Matters. — The total amount of dissolyed mineral mat- ter in any drinking-water depends upon the character of the soil with which the water has been in contact, upon the length of time of expos- ure, and upon the amoimt of carbon dioxide held in solution. Isot even the hardest and most insoluble rocks wholly escape the solyent power of water : no mineral is absolutely insoluble. Silicate of aluminum, which is least acted upon, is soluble to the extent of about 1 part in 200,000. Silicious rocks in general are attacked only very slightly, while limestones are dissolved with comparative ease, and yield con- siderable calcium and magnesium carbonates, especially if the water is rich in free carbon dioxide. Gypsum also is acted upon very freely. Some waters contain very large amounts of mineral matter, derived from deeply situated natural deposits. The Carlsbad springs, for ex- ample, are said to bring annually to the surface enormous amounts of sodium chloride and calcium carbonate, besides 2,500 kilos of calcium fluoride, 600,000 of sodium carbonate, and 11,000,000 of sodium sulphate. Besides the ordinary salts of the alkalies and alkaline earths, most natural waters contain at least very minute amounts of iron. Appre- ciable amounts of iron make water unsuitable for general domestic and technic purposes. It causes staining of clothes if used in the laundry, and headache, dyspepsia, and constipation if used habitually for drink- ing. It cannot be used for dyeing, and as little as 1 part in 1,000,000 makes it unsuitable for use in bleacheries. A quarter of a grain per gallon is sufficient to impart a distinct chalybeate taste. The permissible total amount of dissolved mineral constituents cannot be stated, but 50 parts in 100,000 are generally held to be excessive. Hardness. — Hardness is the capacity a water has for decomposing soap. It depends on the amount of salts of ]Mg and Ca in solution, and hence upon the character of the soil with which the water has been 332 WATER. in contact. Water from rocks which yield linie and magnesia will probably be hard, while that from those composed of alumina, silica, etc., will probably be soft. Some sandstones will yield soft and others hard water, according to the nature of the cement which binds the grains together. The elements causing hardness, particularly the cal- cium salts, have the property of making new combinations with the fatty acids of the soap, and preventing the formation of a lather until thev have been satisfied : 1 grain of chalk, for instance, will use up 8 of ordinary soap before any etfect can be produced ; hence enormous waste of soap occurs from the use of hard water. Hardness is divided into " temporary " and " permanent." The former is due to salts which are removable by boiling ; the latter, to those which are not thereby affected. AA'ater containing considerable free CO^ can take up and hold con- siderable carbonate of lime by means of this gas. Some claim that the carbonate is changed to bicarbonate, but this compound has never been isolated. If the gas be expelled by heating, the solvent power no longer remains, and the amount so held is precipitated, and then can exert no more influence in causing hardness. The chloride and sul- phate of ciilcium are not atfected by boiling. Magnesiimi carbonate is precipitated, but redissolves on cooling. The difference l)etAveen the original hardness and the hardness remaining after boiling is the " temporary " hardness. Permanent hardness is, then, due to those salts not affected by boiling, that is, to calcium sulphate and chloride, and magnesium salts ; and if above 5 parts in 100,000, is commonly regarded as excessive and injuri- ous. Calcium sulphate is not alone objectionable in drinking-water, but also in water used in boilers, since it is less soluble in hot than in cold water, and thus forms a " scale." Scale is of two kinds : that due to the temporary harchiess, easily removed ; and that from CaSO^ which is hard, ver}^ adherent, and removed with difficulty. The latter is deposited the more freely, the higher the tem]icrature of the water. Boiler scale sometimes is due also to other causes. For inst:uicc, A. Reichard ' has reported a case where serious difficulty was caused by the formation of a scale of silicti and lime from a water which contained onlv 2.30 parts of lime and magnesia, but as much as 2.60 of silica. Boiler scale causes great loss of fuel, by interfering with the transmis- sion of heat to the water. Hardness is not only undesiraljle in water used in the laundry and bath, but also in that used for cooking pur- poses, for it makes certain of the vegetables hard and indigestible. Bacteria in Water. The ordinary water bacteria are of the harmless and beueficeut kinds, which, depending upon dead org-anic matter for sustenance, bring about its conversion into simple chemical substances. How many species of these saprophytic organisms exist in water cannot be said, but aliout ' Chemiker iieitimg, 1896, jj. 65. BACTERIA IN WATER. 333 two hundred varieties have thus far been described. They may be present in small or in enormously large numbers without being neces- sarily of hygienic significance, although usually their existence in large numbers indicates the presence of an abundance of organic matter, and jet they may thrive and multiply enormously in water containing al- most no organic food materials. Indeed, multiplication occurs more rapidly in pure than in polluted water, but diminution in nmnber is also more rapid. In impure water, they multiply slowly, but their growth is persistent, and, under ordinary natural conditions, sudden marked •diminution in number does not occur. The ordinary water bacteria are found in much greater abundance in surface-waters than in those derived from the soil. Indeed, many observers, including Koch and Fraenkel, have maintained that waters from the nnpolluted subsoil are practically sterile. This, however, has been shown by Sedgwick and Prescott ^ to be not the case. Using im- proved methods of investigation, and paying special attention to the nature of their culture media, these observers demonstrated conclusively that wholly unpolluted springs, wells, and tube wells may yield consid- erable numbers of bacteria and sometimes a greater abundance than is contained in some surface-waters. In their paper they state " that the plates are remarkable not only for the slow growth of the species present, but also for the absence of liquefying colonies, and, in many cases, for the abundance of chromogenic varieties. These facts are especially important as indicating the total absence of contamination by ordinary surface-water, and, as far as they go, they strengthen the con- fidence with which well-protected ground- waters may be regarded as sources of public water supplies." Their conclusions and results have been confirmed a number of times by other competent investigators else- "vvhere. Ground-waters, when brought to the surface and exposed to the air, soon become rich in the ordinary forms of bacteria, which :find in them the conditions necessary for extraordinarily rapid multi- plication. Surface-waters vary very much m their bacterial content according as the conditions present at any one time favor or retard growth and accessions. Sunshine, influx of food material or of substances inimical to bacterial life, sedimentation, and growth of higher organisms act for or against increase. Suspended matters in their descent carry down with them the bacteria that have gathered upon them or have been entangled by contact. The diminution in their number by this means is more marked in still waters than in rivers with rapid motion. The growth of algse and other water plants causes diminution by removal of the nutrient materials upon which the bacteria depend, and probablv through some other influence not yet discovered. The increase in bacteria, sometimes noticed during the colder months, is explained by Frankland ^ by the fact that in whiter much water runs in over the surface from manured fields. ^ Report of the State Board of Health of Massachusetts for 1894, p 435. ^ The Bacterial Purification of Water, London, 1897. 334 WATER. Besides those forais whose uatural habitat is water, others are often present whose uatural habitat is the bodies of man and animals, and which, in water, are, therefore, in an unnatural medium. These forms, which include the pathogenic varieties, probably do not increase in number in water, whether the latter be pure or extensively polluted. They live for a certain time, retaining their virulence in undiminished degree, and then tend to become modified in this respect and rapidly to disappear. Tho germs of cholera have been found in Seine water in an active state after seven days, and in ordinary drinking-waters as long as twenty days after addition. The typhoid fever organism will live for longer or shorter periods, according to circumstances ; it has been found in very pure water after more than seven weeks, while in badly polluted water its life is very short. Sunshine and temperature appear to have veiy decided influence upon its vitality. The influence of sun- shine is modified by the depth of the water in which the organism is suspended. Buchner ' has shown that the rays of the sun \\ill kill cultui'es of the typhoid bacillus at a depth of about five feet in four and a half hours, while at double that depth their effects are hardly perceptible. While it is tnie that this organism survives longer in cold than in warm Meather, it cannot be said definitely that the rea- son lies in any inherent greater resistance to the influence of cold than to that of heat ; and, indeed, it seems more probable that the explanation is to be found in the fact that in warm weather the con- ditions are more favorable to the growth of the common species of water bacteria which are believed to secrete substances Avhich exert a toxic influence on pathogenic varieties and cause them to disappear. The belief that such toxms are secreted is strengthened by the researches of Frankland,'- who shows that waters which do not favor bacterial multijilication are changed in this particular on being boiled. He found that, while anthrax spores were much diminished in number or actually destroyed in a short time in unsterilized water, their numbers were not reduced and their virulence remained unimpaired in sterile water after upward of seven months. These toxic substances are pre- sumably not secreted l>y all forms of water bacteria, but only by certain species which may or may not be present in any given water, and it is regarded as most likely that they are not inimical to the same extent to all varieties of pathogenic bacteria, but that substances harm- less to one kind may act fatally on another. In general, it may be stated that pathogenic bacteria which form spores retani their vitality and virulence longest in any kind of water. Concerning the significance of B. coli coimnunis, which is exceedingly common in drinkmg- water, there is much difference of opinion. Kruse,^ in 1894, asserted that this organism is so ubiquitous that it cannot be regarded as characteristic of sewage, and in this position he has received the support of a number of other investigators, who have ^ Centralblatt fiir Bakteriologie und Parasitenkunde, XI. )i. 781 ■^ Journal of State Medicine. January, 18Vt4. ' Zeitschrift fiir Hygiene und Infectionskrankh^iten XVII., p. 1. WATER SUPPLIES. 335 succeeded in isolating the organism from all waters examined, although in many cases it was necessary to employ large volumes of the samples. Weissenfeld/ for example, examined thirty waters of good character and twenty-six of bad, and failed in no instance to find it. With bad waters and with some good ones (from deep driven wells, for example) it was found in each cc, but most of the good specimens yielded it only when large volumes (a liter) were planted. Hence, in his opinion, the finding of virulent colon bacilli in water does not necessarily in- dicate fecal contamination. Horrocks,^ ' however, believes that the statement that the organism exists abundantly in all waters and soils is based on a very elastic interpretation of its characteristics, as appears " to be true from Weissenfeld's statement of them. Houston ^ insists upon the importance of paying attention to the relative abundance of this and other sewage organisms in classifying waters as potable or not, and on this point most bacteriologists agree ; but, as Horrocks points out, it becomes a difficult question to decide what quantity of water containing the typical organism is to be considered indica- tive of sewage contamination. From the fact that water, polluted by 1 part of sewage in 100,000, will at the end of two months' storage still show the bacillus if 200 cc. are employed in the test, he would say "that a water which contained B. coli so sparsely that 200 cc. required to be tested in order to find it, had probably been polluted with sewage, but the contamination was not of recent date." Pakes * would regard water, other than that from deep wells, as probably safe whict yields the organism only in greater quantities than 100 cc, but holds that deep well water should not yield it, no matter how much is employed in the test. WATER SUPPLIES. Immediate sources of water supply comprise : 1. Stored rain. 2. Surface-waters, including rivers, lakes, and gathering basins. 3. Ground-waters, including wells, filter galleries, and springs. 1. STORED RAIN. Where other water is not obtainable, and where the natural water is unfit for drinking or for washing and other domestic purposes, stored rain-water is used. If this is collected under proper precautions to prevent the presence of extraneous matters of undesirable character from the receiving area, and properly stored, it constitutes a most wholesome supply. But excepting where rainfall occurs with regu- larity and frequence, the uncertainty of supply, especially in periods of drought, acts as a great drawback. An inch of rainfall is equivalent to 5.61 U. S. gallons per square yard, or 27,152 gallons per acre, but ^ Zeitschrift fiir Hygiene und Infectionskronkheiten, XXXV. (1900), p. 78. '^ An Introduction to the Bacteriological Examination of Water, London, 1901. ^ British Medical Journal, December 21, 1900. * Public Health, March, 1900, p. 385. 336 WATER. only a small proportion of this falls upon surfaces (roofs, etc.) from which it may be collected. The total collecting area of the roof of any building depends not upon the shape and style of the roof, but upon the amount of ground occupied by the building. Thus, a house 40 feet square will have practically 1,600 square feet of watershed, or, allowing for the projec- tion of the eaves, somewliat more, and this whether the roof be flat, pitched, gambrel, mansard, or irregularly disposed. Upon such an area, 1 inch of rain will yield nearly a thousand (997) gallons. The mean annual rainfall of Massachusetts is 43.17 inches, and on this basis, a roof of this size would receive in a year over 43,000 gallons, which would allow for all the needs of the occupants, for drinking, cooking, bathing, laundry, and other purposes, nearly 120 gallons per diem. But under ordinary conditions of storage in cisterns, a very large amount of loss occurs through evaporation, and thus the daily allowance would fall somewhat below this figure. In collecting rain from roofs, it is very necessary to insure cleanli- ness of the supply, by allowing the first flow to run to waste, thereby avoiding contamination by dirt, leaves, bird-droppings, soot, and other matters deposited upon the roof and collected in the gutters. A num- ber of automatic devices are in use for the purpose of diverting the first washings away from the conductors. After this has been done, they change position, so that the subsequent fall is saved and stored. Irregularity in })i'ecipitation is, as has been remarked above, a serious drawback to reliance ujion rain as a sole supply. Partly OMing to a general belief that great battles, in which large quantities of explo- sives are used, are commonly followed by heavy rain, numerous experi- ments have been tried toward breaking drought by discharging power- ful explosives in the upper strata of the atm()s]ihere, but without success. As a matter of fact, the idea of connectictn between battles and rainfall is by no means new, and has, indeed, come down from times antedating the use of gunpowder in warfare. Furthermore, in- vestigation of government records has shown that the popular belief has no foundation in fact, and that great battles have been as often followed by ])eriods of fair weather as by days of storm. Rain-water requires no aeration, for in its descent it has absorbed considerable air ; but melted snow and ice should be shaken with air or poured repeatedly from one vessel to another, in order that they may lose the flat taste so characteristic of unaerated Avater. ^Moreover, their use in the flat condition is believed to conduce to gastric derangement. Snow-water is usually more impure than rain, because the snowflakes, by reason of their larger surface, are more effieient in removing dust and dirt from the air. Cisterns for storage of rain should be so constructed and arranged as to admit of easy inspection and cleansing. They should be kept covered, so as to exclude dirt and dust of all kinds, insects, mice, and other animals, and to shut off light as well, for the presence of light is an important aid to the development of lower plant forms. The SURFACE' WATERS. 337 best materials for their construction are bricks, stone, cement, and slate. Cement makes a good lining if one is desired ; mortar, how- ever, is objectionable on account of the solvent power of water jipon lime, which will cause progressive increase in hardness. Cisterns should be provided with overflow pipes discharging into the open air rather than into the house sewer, and their exits should be protected by wire netting against the entrance of leaves and small animals. 2. SURFACE-WATERS. For public supplies, especially of large communities, surface-waters, as rivers, lakes, and collecting basins, are generally more available than ground- water s . Large rivers and lakes are, unfortunately, very commonly subject to most extensive pollution by sewage of large communities and manu- facturing establishments along their borders, and by the waste products discharged into them from sailing vessels and steamships. Many rivers are subject to progressive increase of pollution by reason of serving as the most convenient receptacle for the sewage of a succession of towns and cities located at intervals from the source to the mouth. Thus, one town takes its water from a point above and discharges its sewage at another place below ; a second, farther down, takes the already contaminated water, and in its turn discharges its sewage at another convenient point, and so on for the rest of the course. On account of the dangers attending the use of such waters, some process of treatment is imperatively demanded to remove the objectionable elements. The different processes available for this work are considered elsewhere. The public mind is being awakened gradually to the wrong practised upon one community by another by the discharge of untreated sewage into what is its only available water supply. In the case of cities located upon the shores of the Great Lakes and other large bodies of fresh water, it is commonly the case that the intake of the water sup- ply is located at no very great distance from the outfall of the main sewers. Smaller rivers and lakes may be subject to the same influences, though in lesser degree ; but, in general, it may be said that these are controlled more easily, especially when they lie wholly within the jurisdiction of a single law-making power. Basins for the collection and storage of rainfall and surface-waters are constructed by throwing a dam across a valley or other convenient depression. Experience has taught, that, even though involving large expenditure, it is best to strip off the surface layers in order to get rid of all organic matter and vegetation, which, if left in place, may prove fruitful sources of trouble. The water which gathers in them has op- portunity to rid itself of much of its suspended matters by sedimenta- tion, and is more often used without further treatment than otherwise. All surface-waters contain more or less active vegetation, and on that account should always be kept exposed to light aud air, otherwise 22 338 WATER. the minute plants will die, and in their decomposition give rise to unpleasant odor, appearance, and taste. Storage reservoirs should have sufficient depth to prevent the water from becoming heated to an unpleasant degree during the warm months of summer. In shallow reservoirs, this is found to be a common occurrence. All sources of surface-water for public supply should be carefully guarded against sewage contamination. It is often necessary to secure protection from pollution by taking great tracts of land and keeping them free from human habitations and industrial plants. 3. GROUND-WATERS. Some large communities and many small ones where no suitable bodies of surface-water are available for public supplies, and the major- it}^ of thinly settled districts which do not admit of public waterworks, dejiend upon the ground-water as the source of supply. For public dis- tribution, the water thus derived is stored in suitable reservoirs, which often must be covered, in order to exclude light. Ground-water is destitute of plant life, but is generally more or less rich in mineral constituents — nitrates, and lime salts, for example — which constitute appropriate plant food. If exj)osed to air and light, vegetable growth may start up and become very luxuriant, and give rise to unpleasant tastes and repulsive odors, while exclusion of light and air prevents the difficulty. For individual domestic supply, storage is not ordinarily necessary, the water being obtained only as immediately needed or pumped periodically into small distributing tanks. In general, unpolluted ground-water of not excessive hardness is preferable to surface-water, on account of the greater exposure of the latter to the many risks of pollution. But it should be borne in mind that all sources of supply, both surface- and ground-waters, may, under one condition or another, be subject to }>olluting influences, and that the conditions prevailing in one locality are likely to be quite diffi?rent from those in another. Ground-M'ater is obtained from springs, or by sinking wells, or by constructing Alter galleries. Springs are merely local outcroppings of the water-table, and are very subject to variations in the volume of outflow. In time of drought, they sometimes cease their flow oimplctcly, because of fall in the level of the ground-water ; and this may ha}i|K'n even in the case of those locatetl at the foot of high hills or mountains. The popular mind endows springs with a remarkable and unvarying degree of purity, but they share with other waters the likelihood of becomuig polluted. The possibility of contamination after and even at the point of issuance from the ground is too often overlooked. Springs are conmiou to some localities and rare in others of similar contour, their presence or absence being determined by conditions not of the surface, but of the geological formations below. In Figs. 25 and 26 are shown in profile two depressions having the same contour. GROUND-WATER. 339 but with very different arrangement of the underlying strata. In Fig. 25 the formation favors the outcropping of springs ; in Fig. 26 the opposite is the case. Fig. 25. Fig. 26. Wells may be classed as dug, driven, and bored. Sometimes they are divided also into deep and shaUoio ; but these terms as a basis of classification are of doubtful utility, since there can be no general agree- ment as to the line of division between them, and because of the absence of any necessarily distinctive peculiarities in the water yielded by ordinary wells of different depths. It is not uncommon to meet wdth general statements that the water of shallow wells is dangerous to health, and should, therefore, be avoided, and that all shallow wells should be condemned and filled. As will be seen, however, shallow wells are not necessarily dangerous, nor are deep ones always safe by reason of mere depth. By some writers, the term deep is applied to wells which obtain their water from below the first impervious stratum, through and beyond which they have been extended ; while the term shalloic is applied to those which draw from what we designate as the ground-water ; that is, that collected over the stratum above mentioned, regardless of the depth at which it lies. With these meanings, it follows that a shallow well may extend farther downward than another classed as deep. The ordinary dug well is a hole dug in the soil down as far as is necessary to reach water, and lined with brick or stone, or, better, with earthenware tubes of large diameter made for the purpose in short lengths with bevelled edges to secure good joints. All brick and stone linings should be well bedded in cement, except near the bottom, and should be faced with the same material throughout their upper part. The impervious lining is necessary for the prevention of the entrance of surface washings ; but it is very generally the case, in some parts of 340 WATER. the country at least, that the ^vell is lined simply with field-stones, without cement, not for the purpose of insuring freedom from surface impurities, but to prevent the sides from caving in. With a proper lining, no surface-water can enter until it has passed through a depth of soil sufficient to insure proper filtration and purification. A dug well should not be left open, but should be closed completely against the entrance of dirt, leaves, and animals, such as toads, moles, mice, and rats. The cover should be supported on a well-set curb, and be sufficiently tight to prevent the return of water spilled or allowed to run to Avaste. A manhole Avith a trapdoor should be provided as a means of inspection and cleaning. For bringing the water to the surface, pumps should be used, and not buckets worked by windlass or well-sweep. In country districts it is a common practice to employ buckets made from kegs, originally used as containers for white lead. It is hardly necessary to call atten- tion to the injury which may be caused by the use of such vessels. The pump may stand directly in the well or away from it and con- nected therewith by means of a ])ipe running laterally and downward. The latter is the better way, as any water wasted at the pump is pre- vented by location, if by nothing else, from running back into the well, and, moreover, the covering of the well, if of wood, is not continu- ally subjected to wetting, which promotes its decay. The best form of pump is the simple lifting ])ump, made of iron or of wood, and con- sisting of an evenly-bored barrel, closed at the lower part by a valve opening upward, and a piston containing another. The upward stroke of the piston, by producing a vacuum, causes the water to pass through the lower valve, and its downward stroke forces the water confined in the barrel through the upper valve, and then the succeeding strokes lift and discharge it continuously. The old-fashioned chain pumps cannot be used without more or less chance of exposure to contamination from above. The action of the wind is very commonly emphiyed as a labor-saver for pumping water not only from the Avell, but upward into reservoirs and distributing tanks. For this purpose a variety of wind-mills have been put upon the market. There are also a number of makes of hot-air engines that are veiy efficient and not unduly expensive. Driven wells, otherwise known as " Norton's tube Avells," " Ameri- can," and " Abyssinian " wells, are made by driving iron tubes of a diameter varying from IJ to 4 inches, according to the needs of indi- vidual cases, into the ground until Avater is reached. The first length driA'en in is provided Avith a jiointed ]ierforated foot, through Avhich the Avater enters the tube. AVhen this length is driven sufficiently far, an- other is screAved to it and the driving is continued, additional lengths being screwed on as necessary. When AA'ater is reached — and tliis is ascertained by means of a Aveighted string let doAvn inside the tube from time to time — a pump is ajiplied and tlie Avater lifted. The first that comes contains sand or fine gravel and dirt, and as this is more GROVND-WATEB. 341 and more removed from below, a pocket is formed which constitutes an undergromid reservoir. Fig. 27. t^ \ , ^ \ '> '.'■''■■',■ * ' • ■^ ' ^ '■■:■ --■■■■ ■". v" ' "' ' \. i Pe rv LOU'S -■ ■? ; J Layer \ ' ^ *■ -v: ;;:: ;f;.\ : -.'" • ^ , ". ■' --'-^ ■ 'V, -' • G-ro u / / ■ '''i '- n'ater V; • Norton tube well. Bored wells diifer but little from tube wells ; in fact, they are prac- tically the same except in the method of their making. They are drilled or bored through solid rock and other strata, and are lined or not with iron pipe, backed with cement according to circumstances. Their cost is much greater than that of the ordinary Abyssinian well, since the labor required is much greater. Sometimes it is necessary, after proceeding several hundred feet with no results, to resort to blasting at the bottom, so as to shatter the rock and form waterways to the well. It is self-evident that wells of these two kinds last mentioned can- not, under ordinary circumstances, become contaminated with surface washings. Both forms are used very commonly not only for individ- ual, but for public, supplies. In the latter case, they are driven in groups, or " gangs," the size of which varies according to the amount of water required. Increase in demand should be met by extension of the system rather than by over-forcing, for the latter will cause an undue lowering of the water level and tend strongly to bring water downward from the upper strata at such a rate as to preclude the puri- fication which normally is brought about by the saprophytic bacteria of the soil. In the case of an ordinary well, the bottom should be considerably below the level of the ground-water, so that when this falls, the well will not run dry, and also because the farther the withdrawal by pumping carries the level of the well below that of the water-table, the 342 WATER. faster will be the flow toward the well, and the greater the supply immediately available. But deepening a well for the purpose of in- creasing the supply sometimes has the very opposite effect, and may even cause it to run practically dry. Suppose, for example, the im- pei"vious layer is underlaid by a thick stratum of coarse gravel, and in the process of deepening the well this stratum is entered : instead of an increase in the supply, it then may happen that the water flowing into the well finds a ready exit downward by the force of gravity into the interstices of the gravel, and the usefulness of the well is terminated. (See Fig. 28.) Included imder bored wells are those known as Artesian. These are bored through impervious strata until a stratum is reached in which the water is under hydrostatic pressure sufficiently strong to force it to the surface, or at least to a point nearly as high, the rise Fig. 28. ■ Per"v i o zi^s S^ rct't'H 'iTv: 1V^\tGr-'. :■ • -— ■■\-''- •■. 'Z-^.z/^. /■■ •o^f /I <:f >x ,V^ W^'t^ir How a well may be spoiled by being deepened. depending upon the height reached by the water-bearing stratum in higher land elsewhere. In Fig. 29 is shown a formation flivorable to the obtaining of water by means of this class of wells. The water in the soil above the first layer of clay may be reached by sinking wells of the ordinary kinds. Below this is a second su]i]>ly confined between two impervious strata inclining upward. The higher this formation extends above the level of the outlet ^ of a well sunk into it at that point, the greater will be the pressure at B and the higher the rise of the water. Thus, if it extends upward to C, for example, the water will not simply fill the tube, l)ut will be thrown some distance into the air. In some cases, although the head developed is very con- siderable, the water does not come to the surface, because of the extent of leakage into the upper pervious strata of the soil. Sometimes the wells are connected with true underground rivers, and sometimes with a]iparently inexhaustible reservoirs which ]ia\T held the water in storay-e for aijes. Sometimes thev derive their water G BOUND- WATER. 343 from fissures draining away the water of surface rivers and lakes, as is proved by the occasional occurrence in the overflow of small fish with eyes. Artesian wells have been known in China and Egypt from very ancient times, and centuries ago they were introduced into the prov- ince of Artois (Artesium), from which their name is derived. They are exceedingly numerous in the western and southwestern parts of the United States, where they have produced enormous results in convert- ing arid, waste lands into fertile farms. Some of them are exceedingly deep, and pass through stratum after stratum of different formations before water is reached. Since the temperature of the earth increases 1 degree Fahrenheit for about 55 feet of depth, it follows that water from these very deep wells Fig. 29. Geological formation favorable to the obtaining of water by means of artesian wells. is materially warmer than that from the upper subsoil. Distinctly hot water from deep sources is rarely fit for ordinary domestic purposes, because of the large amount of mineral matters present in solution by reason of the greater solvent power of water when hot than when cold. Thus they acquire an abundance of salts, which, taken into the body, influence its functions and act as medicines. The presence of organic matters is of importance on account of their reducing power. The sulphuretted hydrogen so common to mineral springs is due to the action of these matters on sulphates. Irrespective of the changes wrought by increased temperature, the water yielded by this class of wells varies very widely in charac- ter. It may bear no resemblance whatever to the other waters of the same district, nor is there any reason why it should, for the con- ditions at the surface and at points hundreds of feet below are quite 344 WATER. different. Moreover, one cannot know how far the water has travelled from where it originally entered the soil to the point where it makes its escape. Of waters from four such wells sunk within the limits of the city of Boston to depths of from 870 to 2,503 feet, two were extensively im- pregnated with common salt and other mineral matter, one was very rich in both vegetable and mineral substances, and the fourth was rich in both these and sulphuretted hydrogen. Drainage Area of Wells. — As to the amount of soil which is drained by a well, there can l)e no general rule. It is commonly asserted that the amount drained may be described as an inverted cone, having the bottom of the well as its apex, and a base with a radius equal to twice the depth of the well. But nuich depends upon the nature and configuration of the surrounding soil, and the extent to which pumping is carried. If the soil be sandy and open, the base will be much larger than if it be clayey and close. If extensively pumped, the well \N"ill drain a greater area than if the demands be moderate ; in fact, the amount of water removed by pumping has a greater influence in deter- mining the drainage area than mere depth. But other things being equal, the nature of the water-bearing stratum determines the distance to which the measurable influence of pumping is felt. Pollution of Wells. — In general, it may be stated that, as between wells of dilferent depths, the shallower are more subject to poHution than the deeper, because of the fact that the latter have the advantage of the greater opportunity for perfect filtration through the soil. But both are subject to pollution by unoxidized matters which enter the soil below the upper few feet in which the nitrifying organisms already referred to are found, as, for instance, from leaching cesspools and leak- ing drains. It is a practice only too common, even on estates of con- siderable size, where the excuse of limited area cannot obtain, to locate the well and the cesspool very near together. To avoid the necessity of having to remove the contents of the cesspool as occasion demands when this receptacle is made water-tight, and to avoid the expense attending this kind of construction, the bottom is generally left open, so that the house sewage may drain away into the surrounding soil. Connection between the cesspool and the well may take . considerable time or may occur quickly, but, once established, contamination goes on uninterru]itedly. Often it happens that the direction of the flow of filth through the soil is wholly away from the well, and contamina- tion may never occur ; but this is a point that can never be determined in advance. It is a common l^elief that, if the well is located in higher ground than the cesspool, there can be no danger of pollution of its water. This, however, is a most fallacious proposition, for it is not so much the location of the outlet of the well that determines the possibility of ])ollution, as the relative position of the cesspool and the point where the Mater enters the well. In Fig. 30 is ilhistrated the manner in which the supply yielded to a pump placed at a point considerably GROUSD-WATER. 345 above the location of the cesspool is polluted directly by the liquid filth issumg from the latter. Again, the geological formation may be such that a cesspool on higher ground than the nearby well will have no influence on the puiity of the water. Thus, a ledge of rock may crop up between them, as shown in Fig. 31, and divert the flow of polluting matters away from the well. Fig. 30. CessPooL Oround/ ■Water How a well located on high ground may be polluted by the contents of a cesspool lower down. In locating wells and cesspools, propert%^ owners not infrequently lose sight of the fact that, while they can govern the disposition of the surface of their respective estates, the conchtions that obtain in the soil below are quite beyond their control. In consequence,^ they may attempt to guard against pollution of their own water supplies by their own excretorv products, without regardmg the possibility of contami- nation bv those of then- neighbors. Fig. 31. How a cesspool located on high ground may fail to poUute a well lower down. The water of newlv dug wells is often of such a character as to lead to the perhaps false conclusion that it is probably polluted by sewage. It is generally turbid, and may, on analysis, yield re.-^ults which, ui case the analvst has not full information concerning it, may seem to warrant a condemnatory report. It may yield figures indicatmg a high content of organic matters, which may disappear as the use of the 346 • WATEE. Avater becomes established. It may even show undeniable evidence of the presence of human wastes, for those engaged in the digging and the stoning may be more interested in the completion of the work than in the perfect purity of the supply, and may be disinclined to go up to the surface for the purpose of relieving the calls of nature. On all ac- counts, therefore, it is better to await the results of a later examination, than to condemn and abandon too hastily a supply, which, within a short time, may prove to be of exceptional purity. Very deep wells may become badly polluted by filth which gains access through open channel-ways, as fissures in rock. • A good ex- ample of this is recorded in the Sanitary Inspector for December, 1896 : A well bored 500 feet into red sandstone drained, through fissures, all the shallow wells in the vicinity. These being of no use as wells, were then utilized as cesspools, and, draining again through the fissures, caused the well to become so foul that it had to be aban- doned. Dr. A. C. Houston ^ shows how deej^ening a well may, in a similar manner, cause its ruin. A well of pure water, 114 feet deep, was deepened by farther boring to 294 feet, when its yield was then found to be imj)ure. At a distance of 800 feet was an old quarry, into which drained the sewage of 25 persons. By fissures in the sand- stone, this reached the water stratum tapped by the extension of the well and thus spoiled the water. On account of the possibility of contamination of shallow w^ells by the entrance of surface washings from above, Koch recommends that pipes be placed in position so as to reach the water stratum, and that then the wells be filled up, first with stone and coarse gravel, and toward the top, for at least six feet, with fine sand. By this pro- cedure, the well is converted really into an Abyssinian well, and is protected from surface contamination quite as well as though it had originally been driven instead of dug. Filter Galleries. — A filter gallery is a large underground tunnel sunk parallel to a river or lake and near to it ; it is in reality nothing more than a horizontal well. The idea which led to their construction was that in this way the river water, percolating outward from its bed through the soil, would be secured in a filtered state, and would ac- cumulate in the underground reservoirs. Although this method of obtaining water has been attended by most excellent results, the fact remains that the Avater so collected comes not from the river, but from the gr» )und on its hither side ; that is to say, it is the ground-Mater intercepted on its way to the river. The water of a river does not, except under unusual conditions, percolate outward, for the silty matters deposited in its flow clog the interstices in the soil of its bed and banks, and act as a valve against its egress. The ground-water, flowing to the river, finds its way in through the silt, which gives way inward ag*ainst the side of least resistance. Thus the silt yields to ingress, and is a bar to egress of water. ' Edinborougli Medical Jounial, Nov., 1894. CLASSIFICATION OF WATERS. 347 The fact that the flow of ground-water is toward rather than away from rivers and other large bodies of water is well shown by the fact that fresh water is obtainable from wells sunk in close proximity to high-water mark on the sea-coast. Such may be not even slightly brackish, although sometimes they are distinctly so by backward diffu- sion of the salts. In the latter case, removal a short distance back- ward obviates the difficulty. That the water derived from a filter gallery is not due to percolation from the river along which it lies, is farther proved by the fact of dif- ference in composition, and especially in hardness. Classification of Waters from the Sanitary Standpoint. From the standpoint of wholesomeness, waters may be divided into two classes : 1. Those free from sewage contamination. 2. Those polluted by sewage. Unpolluted waters are not necessarily suitable for domestic use, pre- senting as they do, wide variations in character. They may be clear, colorless, odorless, and palatable, and contain but little organic and mineral matter ; or they may have high color, turbidity, disagreeable odor and taste, and a high content of dissolved and suspended sub- stances. A water which, by reason of appearance, odor, and taste, due, for instance, to luxuriant growth of algae or other forms of life, is repugnant to the senses, should not be recommended for use, although incapable of producing a specific disease. Such an one requires no chemical analysis to determine its fitness, the evidence of the senses being quite sufficient. Unpolluted waters free from such qualities as render them repug- nant to the senses, and of low content of organic and mineral matters, are suitable for general purposes without regard to their classification as surface- or soil-waters. But, in general, it is held commonly that an unpolluted soft ground-water of good composition is preferable to one of surface origin. Polluted waters may be divided into two classes, according as the pollution is direct or indirect. Direct pollution by sewage is, it is hardly necessary to say, of prime importance, because of the danger of transmission of specific diseases and of lowering the physiological re- sistance of the system. But even direct pollution may be productive of no harmful results, provided sufficient time elapses between the entrance of the sewage at a given point and the use of the water at a distance to permit of the disposal of the noxious elements by natural processes. Thus, a volume of sewage entering the upper part of a large system of public supply may not reach the distributing pipes for several months, during which time its dangerous qualities will have disappeared. Notwithstanding this fact, however, direct pollution of drinking-water should be prevented by all means available, on account of possible risk, and even on aesthetic grounds alone. Indirect pollution is of far less importance than direct. In indirect 348 WATER. pollution the organic matters of the sewage, including bacteria, are filt- ered through the soil, in which they are held hack mechanically and more or less completely oxidized before the containing water reaches its ultimate destination. As to what may be called a safe limit of dis- tance from sources of pollution, no fixed rule can be given : each case must be judged according to its circumstances. The soil as a whole has enormous capacity for purifying water of its contained organic substances and bacteria, both by mechanical retention and by oxidation processes set in motion by the bacteria which inhabit it. But all soils have not this power in an equal degree, and the conditions favorable to its exercise are not always present to the same extent. The soils most favorable for perfect filtration and purification are sandy and gravelly ; in these, the water is exposed in thin layers on the individual grains to the air in the interstices. The latter should be neither too coarse nor too fine. If too coarse, the passage of water is too rapid ; if too fine, not sufficient air can be present at the same time. The organisms are found only in the upper few feet of soil, and it is here also that the con- tained air is richest in oxygen. When the necessary conditions for filtration are present in a given soil, the water which percolates through and reaches the ground-water is quite free from bacteria of any kind, even though the surface is contaminated extensively. Where the soil is very open and permeable to water or fissured, polluting materials may pass through so rapidly that they undergo but slight change on the wav, Avhile Avith a not too fine soil, through which water passes with slowness, purification by bacterial action may be completed within a very short distance. Again, there may be greater safety at a point quite near to, but on one side of a center of pollution, than at another at a considerable distance away on the other side, owing to the direction of the flow of water. Thus, in Fig. 32, the point N, located quite near the point of Fi«. 32. O-0 entrance of the polluting material P into the soil, is far better situated in respect to possible contamination of water by P, than the point D on the other side but fiirther away, since the movement of the water in the soil is, as indicated by the arrows, from >S' toward D, and all impurities entering between the two move from the one toward the other. Similarly, the point S may stand in the relation of point D to some other polluting influence. For the determination of the question whether a given well is receiving pollution from any given point, recourse is had to the diifusi- bility of coal-tar colors, such as fluorescein. An ounce of this sub- stance will impart a very decided color to an enormous volume of PURIFICATION OF WATER. 349 water ; and when it is added to the contents of a leaching cesspool, it will accompany the escaping pollution and reveal to the eye the presence of the latter in any neighboring well-water. Pollution may thus be traced sometipaes through hundreds of feet of fairly close soil. Purification of Water. Before proceeding to the consideration of methods employed to bring about purification of water supplies, a few words are necessary on the subject of " self-purification " of surface-waters. A riyer shows, for instance, at a given point in its course a certain amount of impurity ; at another point farther down, this is found to be considerably lessened ; and farther yet, the diminution is still more marked. This progressive lowering is attributed to the property the water possesses of bringing about its own purification. The practical begiiming of this theory was the assertion, made, in 1869, by the British E,oyal Conunission on Water Supplies, that sewage, diluted twenty times or more by river- water into which it is discharged, will be completely oxidized before it has travelled more than a dozen miles. Two years later, the Rivers Pollution Commission reported strongly against accepting this idea, and concluded that oxidation proceeds with such extreme slowness, even when the polluting matters are diminished very largely by pm'e water, that not only will a flow of a dozen miles not sufhce, but that there is no river in the United Kingdom sufficiently long to accomplish the result claimed. It was then believed that whatever changes occur are the combined result of oxidation and subsidence. It is now recognized that these agencies, assisted by more important ones, namely, dilution, vege- tation, and bacterial action, do in many cases produce very great changes, while in others the results are only partial and of no especial value. Drs. P. Emmerich and F. Brunner ^ showed that in spite of the large amount of sewage matters poured into the Isar in its course through Munich, the water after two hours' flow below the city was practically as pure chemically as it was before it reached it. Jordan ^ has shown that, after thirty-four miles' flow, the Illinois Piver is practically free from sewage bacteria. E. Duclaux ^ has shown the same to be true of the Seine, and other observers have proved it of certain other rivers in England, Germany, and elsewhere. On the other hand, oppo- site results have been obtained by other workers in the same field. Oxidation undoubtedly plays a more or less important part in some, but by no means in all, cases. Dr. T. Meymott Tidy proved experi- mentally that water containing sewage could lose about half its organic matter in from six to nine hours when made to run one mile in glass troughs with abundant aeration. Professor AVilliam P. Mason,* on the other hand, agitated water in a bottle fastened to the connecting rod of a horizontal engine with a ten-inch stroke of 75 to the minute, ^ Die chemischen Yeranderungen des Isar\vassei"S, Munich, 1878. ^ Journal of Experimental Medicine, Dec, 1900, p. 271. ^ Annales de I'lnstitut Pasteur, 1894. * Water Supply, Xew York, 1897, p. 175. 350 WATER. and found that after 9,000 concussions there was but a trifling diminu- tion in the amount of organic matter. A large measure of purification, so far as numbers of bacteria are concerned, is caused by agitation when there are solid particles in suspension in the water. This has been well shown by Percy Frankland,' who observed a decrease of 96.65 per cent, in the number of bacteria. Dilution by access of rain, melting snow, ground-water, and other clean influents, aifects chemical composition favorably, but assists, for a time at least, increase in the numbers of pathogenic and other bacteria. Professor Kebrehl's ^ daily bacteriological examinations of river-water at Prague proved that, in general, the number of bacteria increases with rising water, and is subject to very wide variations, due to a number of causes, among which he mentions changes in the rate of floAv, with consequent alteration of conditions influencing sedimentation, and the influx of temporary pollutions, such as washings from streets and dung- heaps, wliich, under some circumstances, have greater influence than the regular unclean influents. Sedimentation, which formerly was believed to play a very great part in the improvement of river-Avaters, acts to only a slight extent in those which, like the Isar, move swiftly. It is favored by slowing of the current, especially at the river mouth ; and when it occurs it has a very marked influence on the number of bacteria, especially if the water be muddy. This has been shown by Bruno Kriiger,^ who, by a series of experiments, proved that chemically indifferent substances in a state of minute subdivision exert a greater influence, the more slowly, up to a certain limit, they settle ; while other matters, which act both mechanic- ally and chemically, such as lime and hard wood ashes, produce still greater effects. In still water, as small lakes and ponds, sedimentation goes on uno])strnct('d. Bacterial action as a purifying agent is favored by alkalinity and retarded by acidity. It may be important or not, according to circum- stances. Destruction of pathogenic species by the saprophytic class is delayed by dilution by unpolluted water, which, as above stated, favors their increase for a sliort time, after which they ra])idly decline in number. Vegetation was not taken into account by the earlier observers, but has now been ])laccd at the head of the important influences in the process. Pettenkofer * asserted that the greater part of self-purification is due to the growth of algie and other low forms of vegetal)le life, which clean the water of its im])urities in the same way that the higher forms take up and dispose of the manurial matters of cultivated land. This view is endorsed by T. Bokorny,^ who proved that these plants ' Journal of State Modicine, Janiiarv, 1894. '' Bacteriologische iiiul kritische Studien iiber die Veninreinigung und Selbstreini- gnng der Fliisse. Archiv fiir Hygiene, XXX., p. .'52. ■' Die i)liysikalisehe Einwirkimg von Siiikstoflen auf die im Wasser hefindlichen Mikroorganisuien. Zeitsclirift fiir Hygiene, VI I., p. SO. * Zur Selbstreinigung der Fliisse. Arcliiv t'iir Hygiene, XII., p. 269. ^ Ueber die Betheilignng chloropliyllfiihrender Pflanzen an der Selbstreinigung der Fliisse. Archiv fiir Hygiene, XX., p. 181. PURIFICATION OF WATER. 351 take up all manner of organic substances, including volatile fatty acids, amido acids, glucose, and urea. He showed that the water of the Isar contains vast numbers of algse, to whose action much of the changes noted by Emmerich and Brunner were undoubtedly due. Methods of Purification. — The methods employed for the purifica- tion of water embrace : 1. Chemical treatment. 2. Boiling and distillation. 3. Filtration. 1. Chemical treatment is employed to cause the formation of insolu- ble precipitates, which settle out and entangle suspended matters, in- cluding bacteria, in their descent. Alum, for instance, added to the extent of a quarter of a grain to a grain per gallon of natural water containing a moderate amount of CaC03, is decomposed, and forms an insoluble gelatinous hydrate, which combines with the organic matters imparting color and set- tles out as a flocculent precipitate, which entangles the suspended matters, including the bacteria. The sulphuric acid, set free by the decomposition of the salt, unites with the lime or other bases present, and is thus neutralized, and the calcium sulphate thus formed carries down suspended matters in the same manner. If an excess of alum is added, it will necessarily appear in the purified water, and be objection- able on account of its effect on the system, and in the bath and in washing. In case of deficiency in CaCOg, lime-water sometimes is supplied, and identical results obtained. The addition of freshly precipitated alumina serves the purpose equally well, and avoids the presence of the sulphuric acid resulting from the decomposition of alum. Alum removes practically all the bacteria, as has been proved by V. and A. Babes,^ Professor E. Bay Lankester, and others. The use of alum in the purification of water is not of recent origin : it was de- scribed as early as 1830 by Felix d'Arcet,^ who mentions its extensive use in Egypt. Lime-water or milk of lime, added to water containing calcium car- bonate held in solution by carbon dioxide, causes precipitation of the former by uniting with the latter. It thus withdraws the solvent from active service, causes precipitation of that which was held in solution, and, becoming itself converted to an insoluble substance, is precipitated. So a double precipitation occurs. But water thus treated is not neces- sarily limited in its changes to a removal of its excess of calcium car- bonate, for, in the precipitation of this substance, considerable other matter may be carried down mechanically, and bacteria are lessened decidedly in number. Permanganate of potassium is used more or less, particularly in wells in India during the prevalence of cholera epidemics. Enough is ^ Centiulblatt fur Bakteriologie und Parasitenkunde, 1902, XII., p. 45. ^ Note relative a la clarification de I'eau du Nil, et en general des eaux contenant des substances terreuses en suspension. Annales d'Hygiene publique, IV., p. 375 352 WATER. added to secure a slight pink tinge, which indicates a slight excess. This acts as an oxidizing agent Mith good results. For example, Dr. P. W. O'Gorraan ^ relates that during an outbreak at Midnapore, the number of cases, 117, was supposed to have been kept down by its use. These occurred in all parts of the town, excepting in the Euro- pean quarter and at the jail. The former used water whicli was filtered or boiled and filtered, and at the jail especial care was taken of the water supply. Forty -six public and private wells were disinfected with the salt, and the outbreak thereupon ceased. An ounce or ounce and a half or more, according to the size of the well, was dissolved in a bucket, poured into the well, and stirred about. If after half an hour the water showed a red tinge, it was considered that enough had been added ; if not, more was added until a tinge was seen. Accord- ing to Hankin,- enough of the salt to insure a reddish tint lasting twenty-four hours should be added ; but care should be taken not to add so much that fish, -frogs, and turtles, put into wells to keep the water clean, are killed and the water spoiled by their putrefaction. Dhingra ^ states that this method can be relied upon only under cer- tain conditions, and even then its action is not continuous. The agent must ex})end itself first in oxidizing organic matter and nitrites before attacking organisms, which, for their destruction, require it in iairly strong solution. He believes the method to be fallacious in theory, defective m technic, and impossible of practical application. Sodium hypochlorite, ''chlorinated soda," and chlorinated lime, *' chloride of lime," are sometimes used, but in the case of both not without great risk of im]iarting dis;igrecal)le taste. Moritz Traube * gives a simple method of purification by means of the latter agent, which, added to the extent of less than half a gram to a hundred liters, kills all bacteria Avithin two hours. The excess of the agent is neu- tralized bv the addition of somewhat less than half the amount of sul- phite of sodium, which, added somewhat in excess, does no harm, smce it is soon oxidized. The assertion is made that water thus treated possesses no disagreeable taste and has its hardness not appreciably increased. The " Woolf" method consists in adding a 2 to 3 per cent, solution of salt decomposed by a current of electricity of sufficient strength. This is equivalent to adding the sodium hypochlorite itself, Mhich agent, according to Hiinermann and Deiter,^ can destroy in ten minutes all tyj)hoid, cholera, and coli orgjuiisms contained in a liter of water, when enough is emjiloyed to give 40 milligrams of effective chlorine. The sodium compound is more efficient than chloride of lime, for the whole of its available chlorine is almost instantly dif- fused through the water and acts at once. After pm'ification, the chlorine is neutralized by means of sodium sulphite (140 of sulphite to ' Indian Medical Gazette, Julv, 1896. => Ibidem, July, 1896. ^ British Medical .Journal, Aug. 17. 1901. * Zeitschrift fiir Hygiene und Infectionskrankheiten, XVI., p. 149. ^ Deutsche medicinische Wochensehrift, 1901, p. 391. PURIFICATION OF WATER. 353 40 of chlorine), and the water is then practically unaltered in appear- ance, taste, smell, and hardness, but only when the amounts of the •compounds to be added have been most carefully determined. Chlorine as such is used also to some extent ; but, although effective, it is open to the objections that apply to the use of the hypochlorites, of which it is the active agent. Bromine also has its advocates as a chemical purifier, both on a small and on a large scale. Schuraburg ^ recommends a process which is said to kill in five minutes nearly all of the ordinary bacteria and all pathogenic organisms found in water. He uses a solution of 20 parts each of bromine and potassic bromide in 100 of water, 1 cc. of which suffices to sterilize 5 liters of Spree water. After five minutes' contact, the bromine is neutralized with ammonia, and the result is a clear, tasteless, sterile water. Very hard waters and grossly polluted river and marsh waters require larger amounts, because of the presence of lime salts in the former and of ammonia in the latter, which com- bine with the bromine before it has opportunity to act as a germicide. With such waters it is necessary to add enough of the solution to produce a yellow tinge which will persist at least half a minute. In any case, whatever the amount of the bromine solution used, an equal volume of 9 per cent, ammonia water should be added. (In a later ■communication, sodium sulphite is recommended.) This process is recommended particularly for use in the army, and in the tropics, for ships' supplies, and for individual use in times of epidemics. A kilo- gram of bromine is said to suffice to sterilize 16,000 liters of ordinary water. In practice, however, the process has not met with a large measure of success. Schiider ^ has tried the scheme, and finds it unre- liable. It was tested in the Soudan Expedition in 1898, but the difficulties attending its use were enough to lead to its abandonment. Treatment with metallic iron in the form of borings and punchings is employed in a number of places in Europe with most successful results. The best known of the processes in which this agent is employed is that of Anderson, in which the water is delivered into long iron cylinders, on the inner surface of which are curved partial diaphragms which, as the apparatus slowly revolves, carry upward the pieces of iron, which fill about a tenth of the volume of the cylinder, and cause them to shower constantly downward through the water in its passage. The carbon dioxide in the water attacks the iron and forms ferrous carbonate, which, when the water is discharged into the ■open air, becomes oxidized and converted to ferric hydrate. This floc- culent matter entangles much of the organic matters, including the bacteria, and then the whole is passed through sand filters, the effluent from which is very pure and practically sterile. The process is unneces^ sarily expensive, involving as it does, in addition to the first cost of the plant, considerable outlay for power and other items, while the same results in the end may be obtained by the more simple process of sand ' Deutsche medicinische Wochensclirift, March 4, 1897. 2' Zeitschrift fur Hygiene und Infectionskrankheiten, XXXA^II. (1901), p. 307. 23 354 WATER. filtration alone. ]\Ioreover, it appears that witli peaty waters, the organic constituents of Mhieh form sohible compounds with the iron, the results are unsatisfactory. The use of ozone has been recommended as a very efficient method of sterilizing drinking-water, and experiments on a large scale have yielded favorable results. Exjjerimenting on very small (juantities with a Siemens-Halske a])paratus, AVeyl ^ found that 2.3 milligrams of ozone were sufficient to destroy 99 per cent, of the bacteria contained in 200 cc. of water from the Tegel Lake, With 3 and 4 milligrams, he obtained complete sterilization of 0.5 liter of water containing 6,000 bacteria to the cc. For purification on a large scale, the impure water is caused to percolate through a tower filled with pebbles, through which the ozonized air passes upward. The Siemens-Halske apparatus used will produce 20 grams of ozone in an hour. The bacteria are reduced at least 99 per cent, and the percentage of organic matter is greatly diminished, but the process is at present very imperfect, for more than 70 per cent, of the ozone produced is lost. The ozonized water, although free from odor, has an unpleasant taste, and with many persons its use causes derangement of the stomach. This fault neces- sitates further electrolytic treatment with aluminum electrodes, whereby aluminum hydrate is formed and the water is clarified and freed from ozone. Sodium bisulphate has been recommended by Parkes and Rideal ^ in the proportion of 15 grains to the pint. They state that B. typhosus is killed within five minutes, but recommend a contact of fifteen min- utes, in order to insure sterility. Warner^ has found that this is sufficient to cause a striking reduction in the number of added germs, but not complete sterilization. In most cases, B. typhosus is destroyed in thirty, and-7>. cholcrcr in ten, minutes. Contrary to the statement that the agent imparts an agreeable acid taste, Warner finds that to some persons the taste is unpleasant, and to all would probably become irksome. Moreover, a person consuming 5 pints of water in a day woidd swallow 75 grains of the salt, which would tend to increase rather than to quench thirst. Chemical ]iurification of water sometimes occurs without the inter- vention of processes especially provided, of which fact Professor Leif- mann ^ records a conspicuous instance. The Schuylkill River in the upper part of its course receives much refuse mine-water, and becomes impregnated with iron salts and free mineral acids. " In its course of about one hundred miles it ])asses over an extensive limestone district, and receives several large streams highly charged with calcium car- bonate. The result is a neutralization of the acid and a precipitation of the iron and much of the calcium. The river becomes purer, and at its junction with the Delaware River, at Philadelphia, it contain^ ■ Centrablatt fiir Baktorioloyio, XXVI. * Transactions (if the Epidemiological Society, London, XX., 1900-1901. 3 Public Health, .Inly, 1901, p. 700. * Examination of Water for Sanitary and Technic Purposes, Phila., 1895, p. 14. DOMESTIC FILTERS. 355 neither free sulphuric nor hydrochloric acid, only traces of iron, and but a small amount of calcium sulphate. In this manner there is pro- duced a soft water, superior to that of the river near its source, or to the hard waters of the middle Schuylkill region." 2. Boiling and Distillation. — Boiling as a means of purification has been practised from very early times, and, in fact, was advised by Hip- pocrates (460—377 B. c. ; or the avoidance of enlargement of the spleen. This process is quite eilicient so far as destruction of the micro-organ- isms is concerned, but it does not diminish the amount of organic matter. It does, however, reduce the amount of dissolved mineral matter, in that calcium carbonate held in solution by carbon dioxide is precipitated, and calcium sulphate, being less soluble in hot than in cold water, tends to separate out. Boiling is available only to a limited extent ; that is, it is a process w^hich can be carried out in the household, but not on a large scale before public distribution of water. Boiled water is not palatable until aeration has restored the proper taste, but this is easily accomplished by passing it from one vessel to another, or bv ao^itation in contact with air. Distillation constitutes a most efficient process for obtaining pure water. This process produces necessarily a sterile water, which, how- ever, needs thorough aeration. In the apparatus used in the United States Xa\'y, the steam goes to the condensers in company with air, so that condensation and aeration occur coincidently. While no bacteria from the original water can pass over into the distillate, other volatile matters can and do, and instances are common to prove that the dis- tillate of a foul harbor water may produce nausea and diarrhoea in all who drink it. 3. Filtration is a process of purification which is most efficient and available for large water supplies. It is employed on an extensive scale by numerous large cities in Europe and in this country. Before describ- ing the process, however, it is in order to consider filtration in the household. DOMESTIC FILTERS. The domestic filters in common use are, as a rule, useless except for the removal of suspended matters, such as iron-rust, dirt, and other coarse particles, and worse than useless in respect of bacteria, the re- moval of which is claimed but not accomplished, in that they engender a false sense of safety, while they favor the growth and multiplication of organisms. Most of them are small affairs for attachment to a water faucet, filled with a filtering medium of coarse sand, animal char- coal, sponge, ground glass, wool, felt, and other substances which strain out the visible suspended matters not a whit better than the simple flannel bag in common use in New England and elsewhere a quarter of a century ago. They permit the passage of a good stream, and this fact itself is proof of their inefficiency as bacteria filters, for any material sufficiently coarse to permit rapid passage of water is not sufficiently fine to hold back such exceedingly minute suspended matters as bacteria. 356 WATER. Fig. 33. Chamberland-Pasteur filter. Most of the materials used become very foul in a short time, and in consequence the water is richer in bacteria on issuing than it was before entrance. Theoretically, animal charcoal, on account of its oxidizing action, should be an ideal filtering medium, and at first it will re- move a large proportion of tlie bacteria and more or less of any coloring matters. But very shortly it becomes foul ; the calcium phosphate which it contains is of great assistance to the growth of bacteria ; cleaning is impossible, and the effluent, if stored, soon becomes very foul and un]>leasant. The only domestic filters worthy of the name are those which remove mechanically all the bacteria of the water and, at the same time, add nothing of their own substance to the water. Such are the Chamberland-Pasteur, the Berke- feld, and others based on the same principle. In these, the filtering medium is unglazed, well- baked, hollow, porcelain cylinders closed at one end like a test-tube, enclosed within a metallic or glass jacket, with sufficient intervening space for the water, which enters directly from the tap under its usual pressure or " head." The open lower end of the cylinder discharges the water, M'hich passes directly through the walls of the cylinders, or " bougies," in the same Avay in which it would go through blotting-paper. The material is such a very fine straiaer that it excludes all suspended matters whatsoever. (See Fig. 33.) The bougies of the Chamberland-Pasteur filter are made of well- baked kaolin of the proper degree of porosity and hardness ; formerly those of the Berkefeld filter were made of a soft friable infusorial earth peculiar to Germany, called Kieselguhr, but as they were very brittle and very liable to fracture while being cleaned, they are now made of a special blend of clays used in the manufacture of the finest porcelain. Bougies of other makes are of porcelain of varying grades. All these filtering tubes are purely mechanical in their action, and remove none of the matters, poisonous or otherwise, in solution. While they remove and retain on their external surface all the bacteria, they cannot prevent the growth of the organisms from without inward through their walls, and, indeed, this occurs so quickly that, in order to secure absolutely sterile water continuously, it is necessary to clean and sterilize the bougies daily, and thus it is advisable to have two sets, one of which can be cleaned while the other is in use. It has been proved repeatedly that noriual Mater and water artifi- cially and extensively infected will yield on the first day of the use of a clean bougie a perfectly sterile filtrate, and that on the second or third day a very small number of bacteria will most likely be present ; but these are invariably ordinary water bacteria, and if the pathogenic varieties occur in the filtrate, they come considerably later. Repeated FILTRATION OF PUBLIC SUPPLIES. 357 experiments with water infected with B. coll communis, B. typhosus, and B. cholerce have failed to prove the passage of any of these organ- isms, while the ordinary w^ater bacteria go through very readily. To secure a regular supply of wholesome if not completely sterile water, it is, therefore, sufficient to clean the tubes by scrubbing and boiling or by baking about twice a week. It appears, however, that the Chamberland-Pasteur and Berkefeld bougies are not equal in efficiency, for Horrocks ^ has succeeded in growing B. typhosus through the walls of the latter. He attributes this result to the larger size of the lacunar spaces and to the consequently diminished immobilizing and devitalizing influences. Since the shortest time required for the bacilli to traverse the bougie is four days, sterilization by means of boiling water should be carried out every three days, in order to insure complete safety In general, the requirements of a satisfactory domestic filter may be stated as follows : It should yield a sufficient supply of clear, colorless water, free from taste derived from the filter itself; should arrest all bacteria and their spores ; and should be simple in construction, and offered at a low price. Thus far, those made on the principle of the Chamberland-Pasteur filter have met these requirements best. Their introduction into use in the French army in 1889 was followed within two years by a reduction of more than 50 per cent, in the number of cases of typhoid fever occurring therein. Filtration of Public Supplies. Filtration on a large scale is accomplished by the aid of fine sand in filter beds of proper construction, which act both mechanically and biologically. The first beds of which we have accurate knowledge were those constructed by Simpson in London, in the year 1829, which were intended primarily for the removal of dirt and other sus- pended matters causing turbidity. The process was regarded at that time as a purely mechanical one, and though in course of time this kind of filtering medium came into very extensive use, it was gen- erally believed that as carried on there was no marked chemical change in the water, and that what did occur was attributable to oxi- dation of organic matter by air in the interstices of the sand. This was, indeed, the view held generally up to the time when the extensive researches begun by the State Board of Health of Massachusetts in the summer of 1887 proved the great influence of biological agencies, al- though it had been shown by Meade Bolton, Herseus, Plagge, Pros- kauer, and others, that filtration removed all but a trifling percentage of micro-organisms, and that water bacteria exerted some influence on the amount of the usual constituents of water. Although sand filtration of public supplies is of comparatively recent origin, its use for individual house supplies antedates Simpson by at least a century and a half, for Fortius,^ writing in 1685, relates that 1 British Medical Journal, June 15, 1901, p. 1471. ^ De militis in castris sanitate tuenda, auctore Luca Antonio Portio, Vienna, 1685. 358 WATER. the Venetians were accustomed to filter their drinking-water through layers of sand within their cisterns, in order to rid it of disagreeable odor and taste. The first beds constructed by Simpson were broad basins twelve feet in depth, with impervious bottoms and sides, containing layers of stones, gravel, and sand, which occupied half their depth. Beneath the stones were laid ordinary drain pipes, through which the filtered water was discharged. As the top layers of sand became clogged, they were scraped and renewed. The beds of the present day are constructed on very similar lines. They are virtually immense tanks of varying size, shape, and construction. The walls are sometimes vertical, but more often sloping, sometimes built of stone or concrete, and sometimes con- sisting of ordinary embankment. Upon the paved bottom of a bed is laid a system of jierforated or disjointed drain pipes leading to a cen- tral culvert or well, from which the filtered product is drawn. Above the drains are successive layers of coarse gravel, fine gravel, coarse Fig. 34. Partial vertical section of one form of filter bed. sand, and, at the top, one of fine sand from three to five feet in depth. (See Fig. 34.) The fine sand is sharp-grained in character, such as is obtainable at the seashore, and it should not contain clay or other material of similar minuteness of particle ; if present, such should be removed completely by thorough washing. As to the size of the sand parti- cles, it may be stated generally that the finer the grain, the better the effluent ; l)ut, it should be added, the more rapidly it becomes clogged and the more frequently it needs to be scraped otf, and finally, the more difficult it is to wash for future use. AVith the finest sands, the bacteria are removed absolutely, but filtration j)roceeds so slowly that their use is not jwactical^le. The most effective size of grain is a matter on which opinions differ; but whatever the size adopted, it is imjiortant that care be taken to insure uniformity. It is stated variously to be from a fifth to one millimeter in diameter, that is, the diameter of a sphere in volume equal to that of the grain FILTRATION OF PUBLIC SUPPLIES. 359 without regard to the shape of the latter. The higher figure is the one adopted by the authorities at Hamburg. Before the water is applied to the bed, it may be advisable — and if it is from a turbid river, it will be necessary — to allow it to stand several days in a settling basin or reservoir, in order that the sus- pended matters may subside, and thus the too rapid clogging of the interstices of the sand with mud be prevented or retarded. Observ- ance of this precaution will result in lessened necessity of frequent cleaning. Not only are the suspended matters lessened in amount, but organic matters in solution may be destroyed more or less com- pletely by bacterial action, and the bacteria, too, may be diminished in number by being carried down with the settling matters with which they are in contact, and by the death of the less hardy varieties. In the case of waters from ponds and lakes, the preliminary sedimenta- tion proceeds in situ and the settling tank is not needed. The water is delivered continuously at the surface of the bed by devices automatically regulated, and percolates downward through the various layers of sand and gravel to the outlet pipes. Except with very fine sands, the first water of the effluent is not much, if any, purer than the original, but in a short time a sediment layer is formed on the surface aud a slimy algoid gro^^i:h occurs. This super- ficial layer acts both mechanically and by its contained bacteria to cause the removal and oxidation of organic matter and destruction of bacteria. The resulting effluent is quite pure and practically sterile. The Lawrence filter, for instance, removes more thau 97.50 per cent. of the organisms present in the water as delivered, and the reduction is still more marked at the house service pipes, where 99.17 per cent, is recorded, the increase in purification being supposedly due to the fact that their necessary food material has been removed, and hence they cannot long survive. At Hamburg, Altona, Stuttgart, London, and other places, the reduction in bacteria is about the same as at Lawrence. All organic matters are not acted upon to the same extent dm'ing filtration ; some are decomposed very rapidly and mineralized, while others are attacked so slowly that complete removal during the short period elapsing between entrance and exit is often quite impossible. This latter class includes the brown coloring matters so commonly present in surface-waters. These are very stable compounds : they persist during long storage and are nitrified but slowly. In the proc- ess of chemical treatment with alum, however, they are coagulated and removed very quickly. The slime layer, mud layer, or " schmutzdecke," is believed by some to constitute the sole actual filtering medium, the sand beneath acting only as a means of support. But experiments conducted at Lawrence and elsewhere show that this is not true, and that if the greatest care be exercised not to disturb the immediately underlying sand, almost the whole of the slime layer may be stripped off without causing any change in the bacteria count. The greater part of the 360 WATER. work is done in the upper layers of the bed, and yet bacterial efficiency is not necessarily established as soon as a coating has been formed^ A perfectly new filter does not show its best results until it has been in use for some little time, during which the sand particles for a considerable depth become coated with the jelly-like deposit. The active agents in bringing about the death of the bacteria con- tained in the effluent and in accomplishing the destruction and miner- alization of the organic matters are of the same class of nitrifying organisms as are constantly at work in the soil. The death of the bacteria is not directly due to the process of nitrification, for it has been proved that a very marked increase in the process is not neces- sarily accompanied by any diminution in the number of organisms which manage to pass through to the drains. It is possible that the supposed relation of cause and effect is merely a coincidence of con- ditions, that is, that the conditions favorable to nitrification are- unfavorable to the vitality of the ordinary bacteria. It is also {ws- sible that through nitrification the latter are deprived of at least part of the food materials necessaiy to their continued existence and mul- tiplication. Nitrification sometimes ceases suddenly after it has been proceeding for a long time at a proper rate, and then, after an interval, begins again without apparent reason. One explanation offered is that the process begins only "vvhen a certain amount of nitrogenous matter has accumulated within the interstices, that it then proceeds until the store is consumed, and that pending a further accumulation the process lapses. In winter it does not begin again until the tem- perature of the effluent reaches at least 39° F., but after it is once started it is unaffected by a fall to 35°. The most favorable tem- peratures for the process are those of the hot summer months. As to the rate of filtration, it is important that, whatever the rate, it shall be uniform all over the filter. It has been proved by the Massa- chusetts State Board of Health that 2,000,000 gallons ]X'r day can be filtered through each acre of filter bed with the removal of substantially all the bacteria originally present. The Imperial Board of Health of Germany fixes 2,500,000 gallons ]5er acre as the maximum amount per- missible. Koch's three rules of filtration are that the rate of doMU- ward movement should not exceed 100 millimeters an hour, that the filtrate of each section should be examined daily while the bed is at work, and that filtered water containing more than 100 bacteria to the cc. should not be allowed to enter the jiure water reservoir, but should be rejected or refiltered. The bacteriological test is un\v\\ superior to chemical analysis for watching the efficiency of a filter, and a simple count is quite sufficient without attempting to identify the species. AVhen the filter begins to discharge slowly on account of the extent of the algoid growth at the surface, it is not safe to increase the press- ure unduly by flooding the bed with an increased d('])tli of water, for,, as was shown by an experience at Berlin, such a procedure may force FILTRATION OF PUBLIC SUPPLIES. 361 the bacteria, which have accumulated largely in the meshes of the growth, down through the filter at such a rate that they are not de- stroyed by the usual agencies. In this case the water level was raised two feet, with the result that the portion of the city which was supplied with the water of that particular bed was visited by an epidemic of typhoid fever. The same sort of accident occurred at Altoua some years ago, when, a year after successfully going through the cholera epidemic which devastated the neighboring city of Hamburg so exten- sively, a defect in the filter beds was followed by an outbreak of cholera, which disease had then died out in Hamburg. When the sediment layer becomes so thick and dense that with the maximum pressure allowable the required amount of water fails tO' pass, it becomes necessary to scrape off the inch or so that has formed, and then to proceed as though the bed were new. It will require, as a rule, several days for the formation of a new sediment layer, and until it is well developed the effluent should either be rejected or pumped back. The frequeucy with which a bed will require to be scraped depends upon individual circumstances, such as the size of the grains, the character of the water as applied, the rate of movement, the season of the year. The removal of the top is not difficult. It is quite compact and distinct from the sand beneath it, and is readily pared off with shovels or other tools. Successive clean- ings may take place without replacement of the sand, until the depth of the filtering material is reduced to about 15 inches, but not below 12. The scraped-off sand may be washed thoroughly in a machine for the purpose until a sample in a beaker yields no turbidity to clean water, and it may then be stored until needed for future application. Experiments have been tried repeatedly in Massachusetts, Berlin, and elsewhere in sterilizing sand by boiling it in water or otherwise subjecting it to high temperatures, and then determining its efficiency. The results have proved invariably that more bacteria are found in the filtrate than in the original water, and this is explained by the suppo- sition that the bacteria that enter find in the cooked organic matter a food supply most favorable to enormous multiplicatiou, and that the bacteria in the washed sand are necessary for the destruction of organic matter and of some of the varieties of water bacteria. During and immediately after the scraping process, the bed is neces- sarily out of use, and, therefore, it is necessary, in order to insure con- tinuous filtration, to have a number of separate beds, and to scrape them in turn. In this way, while one is out of use, the others can carry on the work. In cold weather, owing to increased viscosity of the water, the rate of filtration is less than in the warmer months. In very cold climates, the formation of thick ice makes proper cleansing of the surface im- possible ; and imperfect scraping causes imperfect filtration. The re- moval of the ice augments considerably the cost of maintenance, and this item alone is one of sufficient importance to warrant the expense of covering the beds. But aside from cost, the efficiency of the process 362 WATEE. is so much greater and the danger of epidemics of water-borne disease is so much diminished that the plant in a cold climate should always be covered. A\'ith an uncovered filter subject to freezing temperatures, imperfect filtration is almost sure to occur periodically, and this is indi- cated by an increase in the daily bacteria count. Thus, AVallichs ^ has noted that after freezing had occurred in the filters of Altona in Feb- ruary, 1886, January, 1887, Februaiy, 1888, and January, 1891-j the number of germs in the filtered water rose considerably, and in each instance, in the following month, there was an lunisual increase in the amount of typhoid fever. Freezing of the surface causes imperfect filtration by bringing it about that the bed is overworked in those places Mhich are still jiervious. The application of water to the frozen surface thaws the ice slowly and unequally, and where the filter is active, it is doing the work of its frozen neighboring areas. Scraping of a bed below the ice cake is performed with a machine which runs between the sand and the ice, cuts the layer and receives it in a bag as fast as it is removed. It is dragged from side to side with- out breaking the ice above it. Covering a filter is advantageous in another direction, for by the exclusion of light, growths of algae are inhibited, and there is, there- fore, less need of frequent cleaning. On the other hand, open filters get the benefit of the sterilizing in- fluence of direct sunlight, but this is more than offset by the promotion of luxuriant growth of algae and other microscopic plants in the warmer months. It is sometimes hardly possible to keep filters in good work- ing order in summer owing to these growths, which clog the interstices very quickly and cause diminished efficiency just at a time when the demand for water is greatest. The coincidence of greater demand and more frequent cleaning does not permit of sufficient intervals of rest after the completion of the scraping process. In Avhat is known as " intermittent filtration," the filter bed is used for the reception of water during part of one day, say sixteen hours, or even during several days, and then is allowed to drain off and rest for a while. As the water drains away, the interstices of the sand become filled with air, that is, the bed becomes aerated, and thus the nitrifs'- ing bacteria which bring about the destruction of org-anic matter and its sul)sequent mineralization to nitrates are assisted to nuiintain their vitality. The intermittent process is superior to the continuous in that nitrification ])roceeds more strongly, the organic matter is, there- fore, more completely removed, and the ordinary bacteria do not sur- vive so long in aerated sand ; but, on the other hand, it is inferior in that, being so much out of active use, the main plant needs to be so much the larger for the accomplislmient of a given amount of work. As a matter of fact, however, all sand filters are at one time or another intermittent, since each time a bed is scraped the water is drained away, and the space formerly occupied by it is then filled with air. Sometimes ' Deutsche mcdioiiiLsche ^Voclienschrift, 1891, p. 25. FILTRATION OF PUBLIC SUPPLIES. 363 it is proposed to put the water through a process of double filtration, that is, to pass the filtrate on to another bed for still further purification. But if the first filtration has been carried out properly, the filtrate will have been deprived of all the materials necessary for the formation of the real filteriug surface on the second bed. Thus the passage of the water through a second filter would be much in the nature of a mere form, for it would pass practically unchanged. kSand filtration, when properly managed, has proved itself so efficient that the number of cities and towns making use of it is growing almost daily. Although protection of a supply at its source may be prefer- able to pollution followed by sand filtration, it is not always so trust- worthy, since pollution may creep in by accident at any time in the l3est guarded supplies. The ideal course is protection at the source, followed by filtration before distribution. This is the method 'now adopted by the authorities of a number of cities in Europe. So great is the importance of filtration and so far reaching is it in its effects on public health, as shown by vital statistics, that the use of unfiltered surface-waters for public distribution is, in Germany, prohib- ited by law. "Mechanical Filtration." — In some places, particularly in^the United States, the water supply is treated in what are known as mechanical filters, of which there are a number of varieties, all based on a common principle. Such a machine consists chiefly of an iron or wooden cylinder filled with rather coarse sand or crushed quartz, through which the water passes by gravity or is driven under pressure at a much faster rate — from 50 to 150 times faster than it moves in a bed. To take the place of the sediment layer which forms in the latter, an artificial film is produced by the use of alum as a coagulant. This is formed quickly and serves the same purpose, though not with the same thoroughness. The filter is called mechanical only because power and mechanical devices are employed in regulating the rate, pressure, the application of the alum solution, and the raking and shaking of the sand in the process of cleaning, which process it is necessary to carry out at short intervals. Instead of removing the top layer, the wdiole body of sand is thoroughly agitated and washed. Filtered water is pumped through from below for five or ten minutes, and the sand layer is agitated by revolving rakes or by compressed air introduced from below. The process is not suited to all water supplies, but for the highly colored and turbid waters so common in the South and West it is particularly well adapted, and is cheaper, more efficient, and more -easily managed than filtration through beds of sand. With careful man- agement, upward of 99 per cent, of bacteria are removed. Plaque Filters. — Another form of filtering apparatus is known as the Fischer plaque filter, which is made of porous bricks about 6 inches (15 cm.) in thickness and about 40 inches (1 m.) square. The porosity is greater than that of the Berkefeld and Chamberland-Pasteur filters, but not very much so. The water is filtered under pressure, and the results are very favorable. At Worms, where both this and sand filtration are 364 WATER. employed, and hence may fairly be judged^ it is found that the two processes are al)out equal in results so far as chemical analyses show, but that the Fischer process passes a somewhat higher proportion of bacteria. The bricks are cleaned by applying a reversed current of water under pressure, but cannot be sterilized like the porcelain bougies, and hence there is likelihood of producing an inferior filtrate by reason of growths occurring in the pores. Removal of Hardness. On account of the enormous waste of soap as well as loss of time which the use of hard waters in washing entails, and of the iujnry to which boilers and hot-water pipes are subject from their action, it often becomes necessary to apply some remedy whereby the degree of hard- ness may be lessened. This may be accomplished by the aid of heat or by the addition of chemicals. Boiling, as Ave have seen, drives off the contained carbon dioxide and causes precipitation of the carbonates which have been held in solution by this agent, but it has no eifect on the salts which cause the permanent hardness. For use on a large scale for public supplies, this means is hardly ap])licable, on account of the cost of plant and of fuel ; but for domestic purposes the cost Ls comparatively slight, in that the fuel necessary in cooking may be utilized coincidently for the purpose of heating water. For the chemical treatment of hard waters, the first process devised was that of Clark, patented in 1(S41. This process is based upon the atfinity of caustic lime for carbon dioxide, with which it forms the practically insoluble carbonate. On the addition of lime water to water containing chalk and mag- nesium carbonate held in solution by carbon dioxide, the reaction occurs, and a double precipitation of the carbonates present and of that formed is brouglit al)out. The jirocess is veiy economical so liir as cost of material is concerned, in that a few cents' worth of lime will remove an amount of hardness which will decompose many dollars' worth of soap. Lime water, however, does not affect the chlorides and sulphates, and hence, like boiling, reduces only the temporary hardness. For the employment of this process on a large scale, various forms of apparatus have been invented, consisting of chambers, or tanks, in which the lime is mixed with water and from which the mixture passes into other large receptacles, Avherein it meets the water to be treated. Thence, according to the nature of the apjiaratus, the water passes on to settling tanks or to mechanical filters, where separation of the precipitate is com]>letecl. The largest plant of this kind in the world is located at Southampton, England, where 2,000,000 gallons of water are treated daily at what may well be regarded as an almost insignificant cost» The building in which it is installed covers less than a seventh of an acre, and is sufficiently large to accommodate additional apparatus whereby its working capacity may be increased by half. Whatever the forms of apparatus em])loyed, the process must be carefully supervised, and the amount of lime added must be constantly regulated ; for if too ACTION OF WATER ON LEAD AND OTHER METALS. 365 little is employed, the full extent of possible softening is not reached, while with too much, the water is made alkaline and the carbonate of magnesium is retained. Caustic soda may be used for softening waters containing carbon dioxide and the salts causing permanent hardness. Added in proper amount to combine with all of the free carbon dioxide, it forms car- bonate of sodium, which, in its turn, attacks and decomposes the other salts and causes their precipitation. Sodium carbonate itself may be added in the absence of free carbon dioxide to bring about the same result. In some processes for softening water, both lime and caustic soda or sodium carbonate are employed, the object being the reduction of both temporary and permanent hardness. Removal of Iron. Some ground-waters contain iron in such amounts as to be objection- able, both on account of its influence on the system and because of its production of stains on linen and other textiles in the laundry. There are two principal methods of removing it, both of which depend upon the conversion of the ferrous compounds into the ferric form, with con- sequent separation as a precipitate. These are filtration and aeration. Filtration may be conducted through sand or coke or animal charcoal, and with either material the iron in solution is exposed to the action of air in the interstices and becomes oxidized to the sesquioxide, which is left on the filtering material. If the air supply is insufficient, and if there is much organic matter present in the water, the sesquioxide may l)e reduced to the ferrous form and again pass into solution. When ground-water containing less than 3 parts of iron per 1,000,000 is ex- posed in large volumes to air, the iron will settle out almost completely within a day or a day and a half. Another method of removal by chemical treatment involves the use •of ferric chloride and caustic lime in the proportion of 1 and 5 to 10 grams respectively to each 100 liters of water. By this, the " Kronke " method, all the iron can be removed, but it necessitates the use of a mixing tank, constant attention, and eventual filtration for the removal of the precipitated iron. Action of Water on Lead and Other Metals. Action on Lead. — The question as to the best material for house- mains and distributing pipes is always an interesting one, and never more so than when a considerable number of persons in a community begin to show symptoms of lead-poisoning, and evidence is presented which incriminates the water supply. Aside from the matter of cost, the advantage of using lead pipes lies in the comparative ease with which lead is worked, since it may be bent to any necessary extent, and thus may be fitted to all manner of irregularities of construction without the need of the frequent cutting, thread-making, and coupling, which the use of inflexible material involves. 366 WATEE. All ordinary waters have a greater or lesser tendency to attack lead, according to the nature and amount of the substances held in solution. The commonly accepted statement, that pure soft water is prone to at- tack lead, and that hard waters tend to protect it by forming incrusta- tions over the exposed surface, is true only in part, for some very pure soft waters exert only very slight action, while some very hard ones act with unusual intensity. Waters containing very small amounts of organic and mineral matters act or not, according as they contain much or little dissolved oxygen or carbon dioxide, or both. A chemically pure water would probably exert no action whatever on chemically pure lead, but commonly neither the one substance nor the other is seen in such a state of purity. Ordinary distilled water, however, which is a nearer approach to absolute purity than any other natural water can be, will, under certain conditions, act very corrosively, the conditions being the presence of the above-mentioned gases. It is held generally that either oxygen or carbon dioxide alone in water has but little influence, but that the two together act with varying inten- sity up to a certain point, directly proportionate to the amount of car- bon dioxide. This belief, based on exjierimental observations of Miiller,^ was strengthened by Drs. Antony and Benelli,- who found that the highest results in lead corrosion were obtained by the use of aerated water charged Avith carbon dioxide. Investigating the plumbo- solvent property of a particular water, A Liebrich ^ came to the same conclusion : that the simultaneous presence of air and carbon dioxide favors action, while either alone has no power. Recently, however, a very extensive inquiry into the subject of metallic contamination of water supplies has been conducted by Mr. H. AV. Clark,* chemist of the State Board of Health of Massachusetts, whose results indicate that oxygen is the more actively corrosive, and that either gas can act alone. He employed distilled water, freed in the first place as conipletely as possible from these and all other gases, and then impregnated with known amounts of either or both. Clean bright lead pipe in equal amounts was placed in half-gallon bottles filled with water containing the gases in the jiroportions stated below, then sealed and set aside at a temperature of 68° F. for one week, at the end of M'hich time the amount of lead taken up was determined. The results are shown in the following: table : No. Gases present. Amount of lead taken up (parts per 100,000). 1 2 3 4 Oxygen to saturation Carbon dioxide 4 ])arts per 100,000 Carbon dioxide 20 ]iarts per 100,000 Oxygen y^^j of saturation, CO, 4 parts per 100,000 . • 2,4100 0.4993 0.8935 0.0861 ' Journal fiir j)raktische Cheniie. Series 2, 36, p. 317. ^ Gazetta ohimica italiana, Jan. 21, 1896, p. 275. ' Zeitsciirift fiir angewandte Chemie, 1898, p. 703. * Annual Report for 1898, p. o41. ACTION OF WATER ON LEAD AND OTHER METALS. 367 A speciaieu of lead in a bottle containing water from which the oxygen had been boiled out as completely as possible, and the carbon dioxide removed by barium hydrate, was kept at 82° F. for a week un- changed. At the end of the second week, slight action was discernible in spots on the surface, and analysis showed 0.0774 part per 100,000 of water. X specimen of ordinary distilled water in a bottle with a small air space in the upper part attacked a similar piece of lead pipe to such an extent that it yielded 10.58 parts per 100,000. In this case, the temperature at which the water ^vas kept was 8 1 ° F. Inasmuch as all drinking-water contains more or less air in solution, oxygen is always present in some amount, and sincej furthermore, car- bon dioxide is also generally present, it follows that, unless substances with a decidedly deterrent influence are present, more or less corrosion is to be expected. Numerous instances of chronic lead-poisoning due to water rich in carbon dioxide are on record. At SommerfeJd,^ for instance, where, in 1888, numerous cases occurred, it was found that the very pure water, rich in this gas, dissolved lead to the extent of about 6 milligrams per liter. At Lowell, Massachusetts, numerous cases were observed during the years 1898 and 1899, and it was dis- covered that one source of supply was rich in dissolved oxygen, and that the other, which caused by far the greater number of cases, was rich in carbon dioxide. Professor A. W. Hoifmann believes that a moderate amount of car- bon dioxide lessens corrosion by forming a protective coatiug of car- bonate, but that an excess of the gas dissolves it as bicarbonate. The gas is said also to have no action on lead coated with suboxide. Water containing free acid of any kind attacks lead. Sulphuric acid, which is supposed erroneously to form an absolutely insoluble compound, the sulphate of lead, is particularly active. In the ordinary chemical sense, sulphate of lead is insoluble in water ; but in the hy- gienic sense, it is sufficiently soluble to be capable of producing serious symptoms. This acid is not an uncommon constituent of water in minute amounts, especially in the vicinity of cities and large towns, where it exists in the atmosphere as an impurity due to the combustion of coal. The peat acids also have considerable action on lead, but they are not always present in waters from peaty deposits. Some very brown waters appear to exert but slight action, while others are intensely corrosive. The peat acids are due supposedh' to the growth of certain micro-organisms found in peaty soils, for a neutral sterilized decoction of peat to which a small amount of fresh peat is added will in a short time develop an acid reaction and ability to dissolve lead. Liebrich ^ reports a peaty water poor in carbon dioxide and carbonates which took up 300 parts of lead per 100,000 over night, and more when calcium carbonate was added. The ammonium compounds and the nitrates have been supposed commonly to have a marked corrosive action on lead. That this sup- ^ Deutsche Yierteljahrsschrift fiir oflentliclie Gesundheitspflege, Suppl. XXIV. ^ Zeitschrift fiir augewandte Chemie, 1898, p. 703. 368 WATER. position is correct, has been proved amply by Mr. Clark's researches ; Ijut intense action is manifested only when the water containing them is exposed to air. The constituents of water which tend to bring about corrosion of lead are, then, carbon dioxide, oxygen, ammonia, nitrates, and free acids. The substances which, on the other hand, exert a protective action include chlorides, carbonates, and silicates, and, probably, sul- phates. According to Crookes, Odling, and Tidy, 0.5 grain of silica to the imperial gallon is sufficient to afford complete protection in all but exceptional cases, even when free acids are present ; but certain waters containing consitlerable amounts of silica are known to be corrosive to a decided extent. The protection due to silica may be obtained by allowing the water to flow through broken flint, flint and chalk, or limestone, but such treatment sometimes has the undesirable effect of increasing corrosive power. Sodium and calcium carbonates are very efficient. The bicarbonate of sodium is generally present in those very soft waters which have the slightest action ; calcium carbonate is efficient whether or not car- bon dioxide is present in the water at the same time. Four grains per gallon are generally considered to be quite sufficient to afford protection under most circumstances. As an illustration of the influence of this agent, may be cited the fact that the very pure water with which Glas- gow is supplied has, before its entrance to the aqueduct, a marked plumbo-solvent })r()j)erty, but loses it entirely in its ])assage to the city, owing to contact with this substance. In 1S87, the Mater of Dessau was treated successfully with calcium carbonate. Sodium carbonate is even more efficient than the calcium salt, but is not always equal to the bicarbonate. At Emden, in LS97, treatment with the latter was suc- <'cssful after the carbonate had failed. Inasnuich as the influences for and against corrosion are numerous and conflicting, the surest method of determining whether a given water will attack lead is to ascertain the truth by actual experiment. Regardless of the character of a water itself, it may be said that its action is greater if the lead is in contact with other metals, so that a galvanic couple is formed. Such may occur, for instance, when a tin- lined lead pipe is bent so that the lining is fractured. Then the two metals being in contact with each other in a more or less saline liquid, the lead, being the more easily oxidized, is dissolved. Again, the tin lining may develop weak s]>ots which may become corroded, and as soon as the lead casing is reached, galvanic action becomes established. Lead pipe containing a small percentage of tin will yield more lead to water than will ordinary lead pi])e, especially if free carbon dioxide is present. A new lead surface will yield more than an old one, as is shown by Professor JNIason, who found that the same water, stored for three and a half months in contact with new and old lead, yielded 58.10 and 3.65 parts per 1,000,000, respectively. ACTION OF WATER ON LEAD AND OTHER METALS. 369 Hot water is more corrosive than cold ; and in the case of either, the solvent power is increased by pressure. The result of the continuous ingestion of minute amounts of lead may be nil or the production of more or less marked manifestations of chronic lead-poisoning. From the fact that lead-pipe is in very general use for house-mains and distributing pipes, and that chronic lead-poisoning is, comparatively speaking, a not very common trouble, it seems reasonable to conclude that with the great majority of persons the metal is eliminated with sufficient rapidity to prevent accumulation and cumulative action. In Massachusetts, notwithstanding the enor- mous use of lead for service-pipes, in but few communities has there been any considerable amount of lead-poisoning reported, and in all of these the water-supply comes from driven wells. Occasionally, fatal lead-poisoning is caused. In one such case, reported by V. Schneider,^ the water was very soft (hardness 1.40) and contained 0.95 milligram of lead per liter. After three months' use of the water, a girl of seventeen died with all the characteristic symp- toms of lead-poisoning. Analysis of the organs yielded lead to the extent of 7.5 milHgrams from the oesophagus, stomach, and duodenum, and 24.7 milligrams from the kidneys, liver, and spleen. Action on Iron. — Corrosion of iron is favored by the presence of * nitrates, nitrites, ammonium compounds, mineral and organic acids, chloride of magnesium, oxygen, and carbon dioxide. The latter is especially active, as has been shown by Professor Leffmann and by E,. Petit. The latter,- investigating the cause of the destructive action of water rich in this gas and poor in lime, placed a certain amoimt of iron iihngs in each of three vessels, one of which contained ordinary water, the second contained water through which a stream of carbon dioxide was conducted for several minutes, and the third was filled with water to which sufficient caustic lime had been added to bind the dissolved carbon dioxide and to give an alkaline reaction to phenolphthalein. After a time, the iron m each specimen was determined, with the following results, which prove the great influence of the gas : 1. 3.15 milligi-ams per liter. 2.- 200.60 miligrams per liter. 3. Only traces. Both cast-iron and wrought-iron pipes may be acted upon rapidly unless their inner sui-faces are covered by some protective coating, such as asphaltum, and even then at the joints where the protecting surface is not continuous or becomes detached. Some surface-waters form a protective layer of vegetable matter on the surface of the pipe, and this is far more efficient than artificial applications, and possesses the addi- tional merit of imparting no unpleasant taste. Action on Zinc. — On account of the action of water on plain iron pipes, pipes of galvanized iron, that is, iron coated within and without with metallic zinc, have been recommended. This lining, however, is ^ Gesundheits-Ingenieur, March 31, 1897. ^ Comptes rendus, 1896, p. 1278. 24 370 WATER. corroded very easily, especially if the water contains oxygen, carbon dioxide, ammonia, or nitrates, and the water is made milky by the oxide and carbonate in suspension. Whether or not the zinc compounds occurring in water can be pro- ductive of harm, is a point on which authorities dilier. But at least it must be admitted that they may cause chronic and obstinate constipa- tion, even when present only in small amounts, and that zinc is not a cumulative poison. Dr. John C. Thresh ^ mentions a case of obstinate constipation in a child, due to the use of drinking-water which passed through a half mile of galvanized pipe. Relief followed discontinu- ance of the supply. Gimlette ^ has reported an extensive outbreak of poisoning attributed to water stored in galvanized iron tanks. Of 56 consumerSj 43 were attacked with gastro-intestinal troubles, the symptoms presented being colic, diarrhtea with consequent anseraia and emaciation, and a spurious kind of dysentery. Analysis of the water revealed large amounts of zinc. Analysis of water drawn from galvanized pipes often has revealed very large amounts of zinc. INIessrs. J. A. and E. W. Voelcker^ record an interesting case in which the hot-water pipes of a house supplied by water ])iped a half mile through galvanized iron, were blocked completely by a deposit of zinc. The water was very pure and soft, and contained but 6 grains of total solids per gallon. The deposit contained 64.32 per cent, of basic carbonate and 21.90 per cent, of oxide of zinc. Zinc is sometimes a normal constituent of Mater. ^Myelins ^ found about 0.5 grain per gallon in the water supply of Tuteudorf, and Carl T. ]\Iorner^ has reported the presence of 0.015 part of zinc car- bonate per 1,000,000 in the water of a well near Upsala. This well, which was about fifteen feet deep, had been in use for more than a year. The water was submitted for analysis solely on account of its peculiar taste, and beyond the fact that it yielded zinc, the source of which could not be determined, the results of the analysis were very favor- able. No unpleasant effects had been noted among those who used the water. Two springs in Missouri, according to Hillebrand,^ yield much larger amounts. In both, the zinc exists in the form of sid- phate. The yield amounts to 120.5 and 132.4 parts per 1,00,000, respectively. Action on Tin. — It is sujiposed commonly that tin is unaffected by water, but such is far from being the case. Tin is attacked by water to a considerable extent, although not so readily as the other metals mentioned ; but the compounds formed are, so far as we know, incapable of causing the slightest injury to the system, and this metal ' Water and Water Supplies, London, 1896. 2 British Medical .Journal, Sept. 7, 1901. 3 The Analyst, .July, 189G. * Ibidem, IV., j). 51. 5 Upsala Liikareforenin.sjs F(".rhandlin,?ar, 1898, Vol. III. * United States Geological Survey, Bulletin No. 13. WATER AND DISEASE. 371 is recommended highly as a lining for iron pipes. Its cost alone pre- vents it from supplanting lead for house-mains and distributing pipes. Water and Disease. The use of impure water for drinking and other domestic purposes may be a direct cause of disease, and such water is supposed also to act upon the system in such a way as to lower the resistance of the body to the action of infectious matters ; but it should be borne in mind that the nature of the polluting material is of far greater impor- tance than its mere amount. To maintain that water containing any considerable amount of organic matter, regardless of its character and source, will tend inevitably to produce a general impairment of health, is as great an error as to underrate the danger possible to arise from a small amount of specific contamination. It is quite as improbable, for instance, that the amount of dissolved vegetable matter necessary tO' yield albuminoid ammonia in what may be designated " considerable ''' amount can do any great injury, even when constantly ingested, as that an infusion of tea, Avhich, by the same process of analysis, would yield results which would be startling in comparison, could of itself conduce to the development of an infectious disease. Alarmists may reject as unsuitable for household purposes a water containing an amount of vegetable matter sufficient to give a yello-w- brown color, and accept as sufficiently pure another containing less organic matter and less mineral matter, but with it the micro-organisms, of infectious disease. They go to the extreme of saying that organic matter in solution must "lower the tone," and should, therefore, be avoided, and if asked why this is so, fall back on " general principles " and "common sense," upon which so much illogical, unexplainable theory is based. As a matter of fact, we kno^v that water which is in a sense impure, but not specifically polluted, may be used year in and year out without injur)\ We know, farther, that water containing much less organic matter, but infected with bacteria of certain kinds, is likely to cause disease in at least a proportion of those who use it once, occasionally, or habitually. We know also that a water once specific- ally polluted may, under similar conditions, be polluted again, and in the interval may be of good quality. We know that water containing large amounts of dissolved vege- table matter in process of decomposition or of a definitely poisonous character may produce disturbances of a very serious nature ; that an abundance of minute water plants, as algse, and animal organisms, as infusoria, may produce ill eifects ; that decomposing animal matters sometimes yield toxic substances of great potency, and that excessive mineral matter in suspension or solution is not without its deleterious effects. Therefore, it may be laid down as a general rule, regardless of the fact that all impurities do not necessarily breed disease or undermine the health, that all water containing or likely to contain domestic 372 WATER. sewage, abundaDt growths of minute vegetable and animal organisms, decomposing matter of animal origin, dissolved vegetable matter of an inherently toxic nature or undergoing decomposition, or excessive amounts of mineral matter, should not be accepted as fit for human consumption. Especially should we bear in mind that polluted water which is quite free from disease organisms and toxic matters to-day may contain them in abundance to-morrow. Disorders Connected with Mineral Matter. — It is noticed very commonly that when one changes suddenly from the use of a soft water to another that is quite hard, there follows a temporary disturb- ance of the functions of the digestive apparatus. The most marked effect is usually constipation with occasional diarrhoea. Loss of appe- tite and slight nauscii are not uncommon. The effects are due to the influence of the salts causing ])ernianent hardness. Change from hard to soft water is quite as likely to cause unacccustomed looseness of the bowels from the withdrawal of this influence on the intestinal secre- tions. Just how much of any one of these salts may be said to be dis- tinctly injurious to health is a matter of doubt, but commonly from 10 to 15 parts in 100,000 of water are regarded as undesirable. It has been asserted that the use of hard water is one of the chief causes of stone in the bladder, but such a connection is extremely improbable. How the use of carbonate and sulphate of calcium can bring about a deposit of uric acid, or of oxalate of calcium, or of phosphates in the bladder, can hardly be explained. The fact also that stone is very common in some districts where water is soft, and rare in some others where it is hard, suggests that the cause is to be looked for rather in the individual himself — his food, his metabolism, his habits of life, and, perhaps, hereditary ])redis})osition. Suspended mineral matter, as clay and marl, will often cause diar- rhoea in persons not habituated to its ingestion, and not infrequently in those who are. The disease most commonly connected with mineral matters in water is goitre. That this disease may be produced l\v drinking-water, can hardly be doubted, for it is a well-known fact that in Switzerland and France, for instance, there are wells M'hich yield waters which are used successfully for the intentional production of the disease, with the view to escape compulsory military service. The enlargement is not neces- sarily a permanent disfigurement ; disuse of the water may be followed by disappearance of the swelling, but oftentimes the disease thus inten- tionally acquired persists. The exciting cause has been attributed to the presence or absence of certain mineral suljstances, but the wide variety of the supposed agents is, of itself, strong evidence of the poor foundation upon which the mineral matter theory rests. It is noticed, for instance, that in some districts where the disease is especially prevalent, tlie soil is largely magnesian limestone, and that, as might be supposed, the ground-water is rich in lime and magnesium salts. Therefore, it is reasoned, mag- nesian limestone must be the cause : but there are many such districts WATER AND DISEASE. .373, where goitre is unknown. More than that, the disease is endemic in some quarters where the water is soft and ahnost free from lime and magnesium salts. Again, it has been attributed to the presence of cer- tain salts of iron, but this theory also cannot bear the test, for these may be present where no goitre is seen, and may be absent where the disease prevails. Absence of iodine is another explanation based on nothing worthy of credence. The most probable cause is now believed by some to be an organism which flourishes in the water. The first to jDromulgate this theory were Italian observers, who, in 1890, reported facts of interest bearing on the question, since which time, other observers, jaarticularly in India, have contributed farther evidence of its probable truth. The most striking facts have been presented by Surgeon-Lieutenant E. E. Walters,^ whose observations were pursued in a district in India 2,000 feet above sea-level, with extremely porous soil and a water sup- ply containing but slight amounts of organic and mineral matter, and but minute traces of iron. The inhabitants, who live under the same climatic conditions, but with diflPerent occupations, may be divided into two classes : the native Bhutias and the Sepoy troops from the northwest provinces. The former are carriers and coolies ; they are omnivorous, but, by reason of poverty, mostly vegetarians. Their chief diseases are goitre, syphilis, and malaria. The temporary inhab- itants, the Sepoys, are all vegetarians, and are a healthy lot, practically free from syphilis, and living under excellent hygienic conditions. They had been in the district twenty months. Examination of 169 Bhutias showed that more than 75 per cent, had goitre ; nearly 70 per cent, of those over twelve years of age were afilicted. Of 380 Sepoys examined, 54 per cent, had goitre. The Bhutias say that their goitres increase during the rainy season, and this is borne out by the out-patient register and regimental admission book for 1895. All the British officers, too, had suffered from enlarged thyroids during the preceding rainy season. Their drinking-water was passed through a Pasteur filter ; all other water used was taken as tea or soda. Taking up the several conditions which have been alleged as the cause of the process, he shows them to be not at fault in this particular district. Iron was present in the water in only minute quantities, and the highest degree of permanent hardness was but 3.5. As to lime as a cause, it appears that many of the Bhutias without goitres are great eaters of lime, while of the Sepoys, who never touch it, more than 50 per cent, developed goitres within twenty months after arrival. The theory that the disease is due to carrying heavy loads up and down hills, might satisfy in the case of the Bhutias, but not in that of the Sepoys, who, though not carriers, yet have goitre. Farther, as to age, it appears that 55 per cent, of the children under twelve had no goitres after living there all their lives, or about the same percentage as developed them among the Sepoys after a visit of only twenty months. He believes the disease to be due to an organism of the amoeba type, with a selective power against the ^ British Medical Journal, September 11, 1897. 374 WATER. thyroid or its secretion. For a time the system opposes it, and some- times successfully ; but when the cause overpowers the phagocytic resources of the system, the thyroid enlarges m the effort to combat the poison. Under thyroid feeding (two 5-grain tal^loids daily) the records show a weekly diminution of a quarter to half an inch in the circumference of the Sepoys' necks, but when the treatment ceases, the gland ag-ain increases in size. That is to say, additional resisting power is conferred by thyroid tabloids, which keep the poison in check and allow the gland to recover its normal size ; but on withdrawing the accessory agent, there is diminished resistance and then again an increase in size. Disorders Connected with Organic Pollution. — Ordinary vege- table matter in suspension and abundant growths of alg» and other water plants sometimes cause diarrheal troubles, but, so far as is known, do not cause specific disease. Peaty matters in solution have now and then aj)peared to be connected with intestinal derangement, but we have no absolute knowledge that they actually have been or can be a cause of such trouble. We know only two diseases which we may say with certainty can be carried by water, and they are cholera and typhoid fever ; but it is said commonly that water is a great factor in the spread of diphtheria, yellow fever, malaria, and dysentery. In the case of diphtheria, the weight of evidence is certainly against its being a water-borne disease. There is some evidence of its spread through the use of a common water supply, but in these cases there is usually a common drinking vessel, and probably a preexisting case of the disease among the drink- ers. The diphtheria f)rg'anism cannot long survive in water Avhich is not very extensively polluted. As to yellow fever, the connection is only suspected ; there is no evidence whatever. Outbreaks of dysentery have occurred which seemed to be due to the use of water contaminated with fsecal matters ; but though it may, perhaps, be an occasional cause, it certainly is not a common one. There is plenty of evidence, such as it is, that water from certain marshes may cause outbreaks of fever of a malarial character, and among the strongest cases are those in which large numbers of persons, as soldiers, have been seized at about the same time after using a water of this sort. But it should not be supposed that all marsh water will produce fever, nor that all water from a malarial district will do so, for many communities use marsh water with no harmful result, and the Mater of a malarial country may be used with im})unity some distance away as a jniblic supj)ly. Such, for instance, is the case with part of the supply of Rome. One of the strongest cases showing possil)le or probable connection between water and malaria is reported by Laveran, and it shows also the action of the stomach at rest and at work. A detiichment of soldiers drank at a certain well, and then enjoyed a hearty meal ; another detachment ate first, anaper. By this means, the most highly colored swamp waters are made colorless in a very lew niiiuitcs. Determination of Nitrogen as Nitrites. — Solutions Required. — 1. SuLPHAXiLic Acid Solution (paramidol)enzene-sulphonic acid). — Dis- solve 0.50 gram in 150 cc. of acetic acid (sp. gr. 1.040). 2. Xaphthylamine Solution (o-amidonaphthalene), — Dissolve 0.10 gram, in 20 cc. of boiling water, filter, and add 180 cc. of acetic acid (sp. gr. 1.040). 3. Standard Sodium Xitrite Solution. — Dissolve 0.275 gram of pure nitrite of silver in pure distilled water and add a dilute solu- tion of pure sodium chloride until precipitation ceases. Dilute to 250 cc. and preserve in the dark in an amiier bottle. 4. Standard Dilute Sodium Nitrite Solution. — Dilute 10 cc. of the preceding to 1 liter with pure distilled water and preserve in the same way. One cc. equals 0.001 mgr. of nitrogen as nitrite. Process. — To 50 cc. of water in a Xessler tube, or to 100 cc. in a tube of larger diameter, add 2 cc. of each of the two first-mentioned solutions. If nitrites are present, a pink to a garnet color is developed Avithin a half hour, the intensity of color depending upon the amount of nitrite jiresent. If no change is observable at the end of a half hour, nitrites may be recorded as absent ; if, on the contrary, a C(,»lor- ation is produced, the test may be repeated, and at the same time one or more comparison cylinders prepared. In similar tubes, dilute to the mark with distilled water free from nitrites 0.25, 0.50, and 1 cc. of the dilute sodium nitrite solution, and add to each the proper amounts of the test-solutions. At the end of half an hour, compare the color acquired by the water sample with the standards, and multiply by the pro])er factor, to determine the amount per liter. Since the air of lal)oratories in which gas is burning is very likely to contain traces of nitrites, which are absorlicd nadily l)y water, it is DETERMINATION OF FREE AND ALBUMINOID AMMONIA. 401 well to keep the tubes corked or otherwise protected. A tube left open some hours is almost sure to develop more or less color. The color reaction is due first to the action of the nitrite present on the sulphauilic acid, whereby a new compound (diazobenzene-sulphonic anhvdride) is produced, which is then acted upon by the naphthylamiue and converted into another (azo-a-amidonaphthalene-parazobenzene- sulphonie acid) which imparts the color. Permanent Nitrite Standards. — Mr. Jackson has proposed employing permanent standards for the nitrite determination also. They are made from two solutions, one made by dissolving 24 grams of cobaltous chloride in distilled Avater, adding 100 cc. of strong hydrochloric acid, and diluting to 1 liter; and the other by dissolving 12 grams of dry cupric chloride, adding 100 cc. of strong hydrochloric acid, and dilut- ing likewise to 1 liter. The standards are made up in 100 cc. tubes, 3 cm. {1^") in diameter and 13.2 cm. (ol'') to the 100 cc. mark. The following table gives the proportions of each solution to be made up to the 100 cc. mark : Co. solution. Cu. solution. Part: 3 of X as nitrite cc. CO. per million. 1.1 +■ 1.1 — 0.001 3.5 + 3.0 = 0.003 6.0 + 5.0 = 0.005 12.5 + 8.0 = 0.010 20.0 + 8.0 = 0.015 The method of determining nitrites, as given by Mr. Jackson, is as follows : Fill a 100 cc. Xessler tube with the water to be tested, add 1 cc. of hydrochloric acid (1 : 4), then 2 cc. of sulphauilic acid (8 grams per liter), and finally 2 cc. of naphthalamine hydrochlorate (8 grams per hter with 10 cc. of strong hydrochloric acid) ; allow to stand twenty minutes until the full development of the color appears. If 100 cc. of water develop a color corresponding to the second of the above standards, for example, it contains 0.003 part per 1,000,000 of nitrogen as nitrite. Determination of Nitrogen as Nitrates. — Solutions Required. — 1. Phexoldisulphonic Acid. — Heat together for six hours in a water- bath 555 grams of strong sulphuric acid and 45 grams of pure phenol. Should the resulting compound solidify on cooling, it may be liquefied again in the bath and then poured into a number of small bottles pro- vided with ground stoppers. Then, as needed, one of them may be placed in the bath and the contents liquefied. 2. Staxdard Solutiox of Potassium Xiteate. — Dissolve 0.722 gram of pure potassium nitrate in 1 liter of pure distilled water. One cc. equals 0.1 mgr. of nitrogen as nitrates. Process. — Evaporate 10 cc. or more of the water with 1 drop of sodium carbonate solution to dryness in a small porcelain dish. To the residue, add 1 cc. of phenoldisulphonic acid, which should be brought into contact with every particle by means of a glass rod. Dilute with water, make strongly alkaline with ammonia or caustic 26 402 WATER. potasli, and, finally, make up to 50 or 100 cc. with water. Evaporate measured volumes of the standard nitrate solution, treat the residues witli a like amount of the reagent, and proceed in the same way to make a comparison scale. The addition of the alkali converts the picric acid, formed by the action of the nitrate on the phenoldisul- phonic acid, into the corresponding picrate, which imparts a bright- yellow color, the intensity of which depends upon the amount of nitrate present. The comparison of tints may be made directly in the porcelain dishes or in tubes of the same sort as used in the nitrite determination. The accuracy of the test is diminished by the presence of chlorides in notable amounts, say more than 2 parts in 100,000, but not by nitrites. On this account, ]\Iason recommends the addition of cor- responding amounts of sodium chloride in the preparation of the color scale. The standards made as above do not change on keeping, and hence may be made up in sets and preserved. Determination of Chlorine. — Solutions Required. — 1. Standard Solution of Silver Nitrate. — Dissolve 4.797 grams of pure silver nitrate in 1 liter of distilled water. One cc. of this solution is the equivalent of 1 mgr. of chlorine. 2. Solution of Potassium Chromate. — Dissolve 5 grams of potassium chromate in 100 cc. of distilled water, add nitrate of silver solution, for the removal of any traces of chlorides present, until a red precipitate of chromate of silver is formed. Let stand, and separate the precipitate by decantation or filtration. This solution is to be used as an indicator. Process. — Place in each of two beakers of similar size 100 cc. of water and 5-10 drops of the mdicator. The beakers standing side by side upon a white surface of porcelain or filter-paper, the silver nitrate solution is added to one of them from a burette Httk» l)v little until, in spite of stirring with a glass rod, a faint reddish tinge begins to be perce])tible. This is seen more easily by comparison with the water in the other beaker. The burette reading is now taken, and then a dro]) or two more of the reagent Avill, by intensifying the red color, show that the end ]K)int has been reached. The process depends ujum the fiict that silver has a greater affinity for chlorides than for chrom- ates, and that, so long as any of the former is present, no permanent union will occur with the latter. AMien, however, all the chlorine has combined M'ith the silver, the nd chromate begins to form, and makes its presence known by the change of color. On com))letion of the process, the amount of the standard reagent used indicates the amount of chlorine present, each cc. used representing 1 milligram. Tnasmuch as a certain amount of the reagent is required to give the beginning tint in 100 cc. of distilled water, a correction should be made before setting down the result. This amount differs somewhat with different observers, since all eyes are not e(|ually ijuick to discern tiie appearance of the reddish tint, and hence the best method of fixing DETERMINATION OF FREE AND ALBUMINOID AMMONIA. 403 the amount to be subtracted is for each one to determine it himself by experimenting with 100 cc. of distilled water containing the requisite amount of the indicator. Should the amount of chlorine in a given sample be so small that the end reaction appears on the addition of but a few drops of the silver solution, it is best to concentrate 250 or 500 cc. of the water to 100 cc, and repeat the titration. Determination of Residue. — Evaporate 100 cc. of water to dry- ness in a perfectly clean, dry, accurately weighed platinum dish. When completely evaporated, transfer the dish from the water-bath to an air- bath kept at 105° C, and leave it for an hour, at the expiration of which time, place it m a desiccator to cool. Reweigh and note the gain in weight, which represents the amount of total solids in the vol- ume of water taken. The number of milligrams gained represents the number of parts per 100,000. The weighing should be done as quickly as possible, in order to avoid error due to the absorption of atmospheric moisture by hygroscopic matters in the residue. In order to determine the amount of volatile substances, the dish is next heated to dull redness on a platinum triangle over a Bunsen lamp. The organic matter, in the process of burning off, gives rise to more or less blackening, and may also evolve odors which often convey some idea of its nature. The blackening may disappear quickly or may persist for some time, especially in the case of woody matters, such as are present in brown swamp waters. Animal matters cause an odor like that of burnt horn ; vegetable substances, one suggestive of burning peat. The loss in weight represents not only the organic matter, but also the nitrates, nitrites, ammonium salts, combined carbonic acid, and^ if the temperature has been raised too far, part of the chlorides. The residue after ignition represents the "fixed solids." Determination of Hardness. — For the determination of hardness, a number of processes are in use ; but for practical utility, that known as the " soap method " is to be preferred. Solutions Required. — 1. Standard Solution of Calcium Chlo- ride. — Weigh out 1 gram of pure calcium carbonate, dissolve it in as little as possible dilute hydrochloric acid, and evaporate to dryness. Add to the residue a little distilled water, and again evaporate to dryness. Dissolve in distilled water and make up to 1 liter. One cc. represents 1 milligram of calcium carbonate. 2. Standard Solution of Soap. — Scrape about 10 grams from an old dry piece of pure Castile soap free from sodium hydrate and carbonate, and dissolve it in 1 liter of diluted alcohol. Let stand over night and filter. This should next be standardized in the foUomng manner : To 100 cc. of distilled water contained in a glass-stoppered bottle of about 250 cc. capacity, run in, from a burette, successive small portions of the soap solution until, on vigorous shaking, a lather is formed which persists at least two minutes, and note the amount used. Repeat the operation with 99 cc. + 1 cc. of the standard solution of 404 WATER. calcium cliloride, and then with 2, 3, 4, 5, G, 7, 8, 9, and 10 cc. of the same solution made up to 100 cc. with distilled Avater, and note the amount used in each test. It will not suffice to determine the amount necessary to produce a lather with distilled water and with 10 cc. of the calcium chloride solution, made up to the same volume, and divide the difference bv 10, since as we go up in the scale a gradual lessening of the amount of increase for each degree is noted. In this way we obtain a scale of values for the particular lot of soap solution made at one time. It will save some trouble if one makes up a number of liters, but it is necessary to make occasional tests to see that the strength does not deteriorate, or, if it does, to correct the scale. Process. — To 100 cc. m the bottle above mentioned, add the soap solution in the same manner as employed in making the scale, and, when the end point is reached, note the amount used, and, by reference to the scale, ascertain the number of degrees of hardness. Should the water be harder than 10 degrees, it is best to take a smaller amount and make it up to 100 cc. with distilled water and then proceed anew, remembering at the end to calculate accordingly. The result obtained expresses the '* total hardness." If it be desired to ascertain the temporary, or removable, hardness, 100 cc. of the water may be boiled five minutes and then allowed to cool. The original volume is restored by the addition of the necessary amount of distilled Mater, and then the operation is repeated. The second result indicates the permanent hardness, and the difference, if any, is the temporary hardness. Determination of "Oxygen Required." — All organic substances are susceptible of oxidation ; but as they are widely variable in character, they require very different amounts of oxidizing agents for the attain- ment of the same result. The several methods proposed for determin- ing the oxvgen-cousuming caj)acity of drinking-waters have, therefore, only a limited value ; but, in general, it may be said that a high require- ment indicates an amount of organic matter inconsistent with purity when it cannot be accounted for by the presence of ferrous salts. Since the amount of organic matter is indicated pretty fairly by the ammonia and albuminoid-annnonia determinations, the estimation of the " oxygen required " serves only as confirmatory evidence. Solutions Required. — 1. Standard Solution of* Potassium Per- manganate. — Dissolve 0.395 gram in 1 liter of distilled water. One oc. is equivalent to 0.1 mgr. of available oxygen. 2. Standard Solution of Oxalic Acid. — Dissolve 0.7875 gram in 1 liter of distilled water. One cc. corresponds to an equal measure of the permanganate solution. 3. Dilute Sulphuric Acid, 1 : 3. Pro("ESs. — The determination is based on the fact that potassium permanganate gives up its oxygen readily to organic matter, especially in the presence of acid and with the application of heat. The reaction is expressed in the following equation : 4KMnOi -r 6H,S04 = 2KJiOi + 4MnS04 + eH^O + SO.,. DETERMINATION OF FREE AND ALBUMINOID AMMONIA. 405 Thus 4 molecules of permanganate will yield 5 of oxygen, or, cliiFer- ently expressed, 632 parts by weight of the one will yield 160 parts by weight of the other ; hence, 3,950 of permanganate equals 1,000 of oxygen. In this operation, cleanliness of vessels is of the greatest importance. A porcelain casserole or evaporating dish of sufficient size is made fit for use by boiling it in distilled water acidulated with sulphuric acid, and adding permanganate solution until no further decoloration is observed. Place 100 cc. of the sample in the dish, add 10 cc. of the dilute sulphuric acid, and heat to boiling. Add from a burette sufficient of the permanganate solution to cause a very distinct redness, and boil again, adding the permanganate as the color tends to fade, so as to retain as nearly as possible the original color. When farther boiling for five to ten minutes fails to diminish the intensity of the color, oxida- tion is complete. Add now 10 cc. of the oxalic acid solution, which will discharge the color if the permanganate has not been added too freely. Should the color not be discharged by 10 cc, add 10 more. Having now a colorless solution, add more permanganate until a slight pink color again appears. Note the total amount of permanganate used, subtract from it that used up by the oxalic acid, multiply the number of cc. remaining by 5, in order to arrive at the amount which would be consumed by 1 liter, and divide by 10 to express the result in milligrams of oxygen. Inasmuch as any nitrites present are oxidized to nitrates, a correction should be made for them. This can be done very readily, since 16 parts of oxygen are required for 14 parts of nitrogen as nitrites. Since the permanganate solution is not wholly stable, it should be titrated against the oxalic acid solution every time it is used. This may be done most conveniently by adding, after the operation is com- pleted and the reading of the burette is noted, 10 cc. more of the oxalic solution and titrating to the same point as before. The oxalic solution keeps better, if a few cc. of strong sulphuric acid are added when it is being diluted to 1 liter. Determination of Color. — The color of water may be observed by viewing a sufficient depth of the specimen in a glass cylinder against a white surface. Color may be expressed quantitatively by comparison with the standards for the ammonia determinations. Determination of Odor. — Place about 200 cc. of water in a 500 cc. beaker, cover with a watch-glass, and heat to about 40° C. Give the beaker a rotary motion, so that the water is set in motion, remove the watch-glass, and with the nose well inside the beaker note the character of the odor. Some analysts prefer to heat the water in a glass-stoppered bottle, the use of which permits a much more thorough agitation of the water before applying the nose. The odor should be designated according to the substance which its presence suggests. Determination of Reaction. — A most delicate reagent for alkalinity ia water is a 1 per cent, solution of toluylene-red. Fifty cc. of water 406 WATER. distinctly alkaline will become intensely yelloM' on the addition of 2 or 3 drops. A less degree of alkalinity will cause an orange or pale- red color. It is so delicate a test that 1 part of alkaline carbonate in 1,000,000 is revealed by it. The presence of acids is shown by another sensitive indicator, lac- moid. This is not affected by carbonic acid, nor by ferrous and other metallic salts which are acid to litmus, but is affected by ferric salts. It may be used as a 1 per cent, solution in diluted alcohol. Phenol- jihthalein solution, 0.5 per cent., is colorless in neutral and acid solu- tions, and pink in alkaline. It is affected by carbonic acid. In the determination of reaction, a drop or two of the indicator may be added to a volume of the water in a long glass tube. A very faint change, due to acids or alkalies, is perceptible on looking down through the column against a white background. If the reaction is acid, the sam])le should be boiled, then cooled, and tested again to ascertain if the acidity is due wholly or in part to carbonic acid. Acidity and alka- linity are determined quantitatively by titration with centinormal solu- tions of sodium hydrate and hydrochloric acid, nsing lacmoid or phenolphthalcin and methyl-orange as indicators. Determination of Turbidity. — For the determination of the degree of turbidity, several methods are in use, among which the following may be mentioned : Mason ^ recommends standards made by adding weighed amounts of kaolin to distilled water, each representmg parts per 1,000,000 of kaolin. Whip])le and Jackson^ employ finely powdered diatomaceous earth, instead of kaolin, because of the greater uniformity in the size of the particles. Hazen ^ measures it by determining the de]ith at which a 0.1 mm. ]ilatinum wire can no longer be seen. Detection and Determination of Lead. — Many processes have been proposed for both (|ualitative and quantitative determination of this most undesirable contamination. The simplest test, but by no means the best, consists in adding a drop or two of ammonium sul- ])hide to a volume of water in a tall glass cylinder, and noting the character of the discoloration jiroduced. If darkening occurs, due to the formation of a metallic sul])hide, the addition of dilute hydrochloric acid Mill distinguish between lead and iron, the sulphide of the latter being soluble. To those who have had practical experience in de- tecting minute amounts of metals in water, this method is far from satisfoctorv. More or less color is im])arted by the ammonium sul- ])hide, and more or less turbidity is produced commonly on the ad- aces covered by the culture medium. Drs. W. Hesse and Niedner^ have ])ro])osed a method for quantita- tive determinations which obviates the use of nutrient gelatin and all the attendant disadvantage due to li<|uefying bacteria. Their most im])oi"tant recommendations are as follows : The amount of water planted per plate should be such as will pro- duce no more colonies than may easily and accurately be counted ; that is to say, not more than a hundred. At least five plates should be pre- pared, and if these yield results fairly in agreement, their average should be taken; but if any plate gives figures more than 100 per cent, removed from the mean, it should be discarded. These should * Zeitsclirift fiir Hygiene und Infectionskrankheiten, XXIX., p. 454. BACTERIOLOOICAL EXAMINATION OF WATER. 415 be kept at room temperature and in the dark as long as new colonies develop, that is, for two to three weeks, at the end of which time one may make counts which will have some claim to accuracy. On account of the evaporation that occurs during this time, it is necessary to use for each plate at least 10 cc. of culture medium. The most suitable medium is made with 1.25 parts of agar-agar, 0.75 part of albumose, and 98 parts of distilled water. Gelatin should not be used, on account of the rapidity with which plates made with it become useless on account of the liquefying colonies, which obstruct some growths and dissolve and wash others away. The medium above described possesses the advan- tao-e that it needs no addition of acid or alkali. Qualitative Determination. — The chief interest in qualitative ex- amination of drinking-water lies in the solution of the question whether or not intestinal bacteria are present. Plates may be prepared and pre- served in the same manner as for quantitative work, except that the amount of water planted needs no accurate measurement, but, on ac- count of the usual great preponderance of the common harmless bac- teria, it is rarely the case that one can isolate the pathogenic varieties without recourse to special methods which favor their growth and at the same time kill oif the others. One such method consists in incubating the sample with bouillon for thirty-six to forty-eight hours at 37°-38° C, and then preparing plates in the usual way. The common bacteria are in this way subjected to conditions unfavorable to their growth and vitality, while multiplication of the pathogenic varieties is promoted. Another method, depending upon the resistance of the typhoid organ- ism and B. coll communis to the action of dilute carbolic and hydro- chloric acids and the powerful iniluence of these agents on the common water bacteria, consists in incubating a few drops of the sample in 10 cc. of bouillon containing a few drops of a solution of 5 parts of phenol and 4 of hydrochloric acid in 100 of distilled water. The bouillon is incubated first with the phenol mixture for twenty-four hours at 37° C, and then receives the sample. After farther incubation for twenty-four hours or longer, plates are made in the usual way and the resulting growths systematically studied. Still another process is that of Professor Theobald Smith,^ who recommends the addition of a few drops of the water to bouillon containing 2 per cent, of glucose, and incubating in fermentation tubes at 37°— 38° C. for thirty-six to forty-eight hours, at the end of which time, if gas has accumulated in the end of the tubes, plates are prepared in the usual way. By this process it is possible to secure pure cultures of the intestinal bacteria. For the detection of B. coll in water supplies, the use of neutral-red has been proposed. Rothberger,^ in 1898, found that this color is re- duced by this organism, but not by B. typhosus, and is changed to can- ary-yellow with green fluorescence. Later, he discovered that certain anaerobes have the same power (i>. tetani, B. oed&natis maligni, B. an- thracis sym-ptomatici). The majority of the aerobic pathogenic bacteria ^ American Journal of the Medical Sciences, September, 1895. ^ Centi-alblatt fiir Bakteriologie, etc., XXIV., p. 513. 416 WATER. have been tested with negative results. Scbeffler^ found the reaction to be constant in his investigation of a large number of races of B. colij and concluded that organisms which give negative results may be excluded from the coli group. MakgilP found that B. tctanl and B. oedematis maUgni produce the same effect as B. coli in glucose-agar even when the surface of the medium is exposed to air, but in bouillon the anaerobes produce the reaction only when oxygen is excluded. He discovered that B. mesenfericus changes the red to a dull orange both in bouillon and in glucose-agar. His experiments thus far seem to indicate that a water producing a typical canary-yellow in neutral-red media, within forty-eight hours in bouillon, and accompanied in glucose-agar by green fluorescence aud gas formation, may be considered to contain B. coli. He finds that 1 to 5 organisms per cc. will produce the reac- tion in bouillon within twenty-four hours^ and large numbers within twelve hours. He concludes that neutra'-red affords a rapid and very delicate test of the presence of B. coli ; that, using varying quantities of water, a rough estimate may be obtained of the number present ; that a negative result with a fair sample is evidence of their absence; aud that farther investigation is needed to show that the reaction is positive evidence of their presence, although in his experiments it was present whenever the reaction occurred. Savage,^ using glucose media (broth or ag-ar containing 0.5 j)er cent, of glucose), found that agar gives the best results. He warns against employing an excess of neutral-red, for it may not be reduced. His best results were from 0.1 cc. of an 0.5 per cent, aqueous solution of Griibler's neutral-red added to 10 cc. of broth or ag*ar. From the results of an extensive series of experiments, he concludes that a posi- tive reaction is not absolutely diagnostic of B. coli, but in the vast majority of cases points to its presence ; that a negative reaction does not exclude it, but makes its presence highly im])rol)able ; and that the test is ver}" easy to apply, and of great value in the routine examination of water. For the detection of the cholera organism, Koch recommends incu- bating 100 cc. of the suspected Avater with 1 gram each of sodium chloride and peptone at 37° C, and preparing therefrom at intervals of ten to twenty hours agar-agar plates, Ironi which, if colonies develop, otlier ])lantiugs may be made for systematic investigation. For the detection of the cholera organism in small numbers in the presence of water bacteria. Dr. Arens * has found caustic potash of assistance, and recommends incubating 175 cc. of water with 25 cc. of pancreas bouillon containing from 0.10 to 0.10 gram of the alkali, by means of Avhich the growth of the organism is favored. Dr. A ufrecht,'' working on similar lines, found that the development of the organism is favored by gelatin containing 1 per cent, of caustic soda. ^ Centnilblatt fiir Bakteriologie, etc., XXVIII., p. 109. * Journal of Ilypiene, October, 1901, p. 4IW. ^ Ibiileni, p. 437. * MiinolK'nor inedicinisclie Woohenschrift, March 7, 1S93. * Centnilblatt fiir Bakteriologie und Panisitenkunde, March 23, 1893. COMPARATIVE VALUE OF ANALYSIS OF DBINKINO-WATEB. 417 Comparative Value of Chemical and Bacteriological Analysis of Drinking-water. As the science of bacteriology began to develop and take the posi- tion to which its importance gave it a title, its disciples conceived a strong prejudice against and contempt for any opinion as to the pota- bihty of a particular water based upon chemical analysis, maintaining, quite correctly, that minute amounts of ammonia, albuminoid ammonia, and chlorine are incapable of acting as the exciting cause of infective disease, and that not these substances, but only specific organisms not demonstrable by chemical processes, can so act. It must be conceded that, for a time prior to the discovery of the nature of the infective agents, the importance of the results of chemical analysis was grossly exaggerated, and that arbitrary standards, such as were established by the Rivers Pollution Commissioners, upon which conclusions were based, have, in the light of farther experience, been abandoned as absurd and untrustworthy. But it must also be conceded, even by those who were most caustic in their criticism, that chemistry is to-day equal, if not superior, to bacteriology in indicating possible danger from the use of water exposed to contaminating influences. In the earlier days, much capital was made by bacteriologists of the fact that a sample of water, inoculated with a culture of the bacillus of typhoid fever, was reported by a chemist of high standing as of great purity and eminently suitable for domestic purposes. Such a test, however, is unworthy of the slightest consideration, siuce imder natural conditions a water showing a high degree of chemical purity is not likely to be infected with a pure culture of a pathogenic organism, and the submission of a pure water so treated is a mere trap, the setting of which is no more praiseworthy than would be the sending of a sterile solution of cyanide of potassium or of sulphate of strychnine to a bacteriologist with a request for an opinion from the standpoint of his specialty as to its desirability as a beverage. Chemical analysis can show the presence of organic and mineral impurity such as accompanies infectious matters from the intestine and bladder. It cannot give grounds for a positive assertion that the use of a water thus polluted will inevitably cause disease, but it can and •does serve to point out possible danger. It can detect the presence of sewage matters, and while it cannot prove the presence of infectious material therein, it can at least point out that the occurrence of typhoid fever in the community furnishing the sewage is likely to be followed by other cases of the disease in the community which uses the polluted water. It cannot distinguish typhoid pollution from any other excre- mental contamination, since a healthy body yields the same chemical substances as one that is diseased. In the case of waters containing no evidence of contamination, it can supply the basis of an opinion as to safety, but it cannot furnish any guaranty that the condition is per- manent. 27 418 WATER. Bacteriological analysis differentiates between pathogenic and non- pathogenic contamination, but it is only rarely that it serves to point ont danger in advance. Even when an outbreak of typhoid fever has occurred and attention is drawn thereby to the condition of the water supply, the results of bacteriological examination are generally negative. The reason for this is twofold. In the first place, the examination for the detection of the specific organism is not ordinarily begun until attention is drawn to its necessity by an outbreak of the disease, which does not appear until about two weeks from the time contamination has occurred. Unless the contamination is continuous, by the time the examination is instituted, the polluting materials have either passed on or the specific organisms have perished. In the second place, even although they are present, with our present methods it is not an easy matter to isolate them, and we can determine in most cases only the probability of their presence. It should be borne in mind that the organisms are particulate bodies in suspension in great dilution, and that their distribution is not homo- geneous as is the case with substances in solution, and that, therefore, the amount of water taken for planting plates may not contain them. But in the unsuccessful search, it is not uncommon to find B. coli communis, and where this organism lurks, the other may have been present. As a rule, bacteriological search for the typhoid bacillus has given negative results. Laws and Andrews failed to find it in the sewage of London, although it must have been present ; and they had but slight success in the examination of sewage from a hospital where forty cases of the disease were being treated. The reason for this may be that through absence of suitable food material and favorable temperature, and bv reason of the antagonistic influence of the ordinary sewage bacteria, the typhoid bacilli had lost their vitality ; or it may be that they were so diluted that the volumes used for planting failed, as a rule, to contain them. Examination of the water supposed to be con- cerned in the unusual outbreak at Maidstone yielded absolutely nega- tive results, although no reasonable doubt can exist that at some time they had been present. Professor Percy Frankland, who has had a large experience in dealing with micro-organisms in air and water, says : ' " The detection of specific pathogenic bacteria in drinking-water is now known to be almost beyond the range of practical politics, and the search for such bacteria is, in general, only caiTied on in deference to the special request of the layman, the uninitiated, or the hopelessly ignorant, whilst it can- not be repeated often enough that any feeling of security Avhich may be gathered from an unsuccea«ful search for pathogenic bacteria is wholly illusory and in the highest degree dangerous. . . . By far the most important service which has been rendered by bacteriology is the means which it affords of controlling the efficiency of filtration and other purification processes. The slightest irregularity or defect in the proc- * Jouraal of the Sanitary Institute, October, 1899, p. 393. COMPARATIVE VALUE OF ANALYSIS OF DRINKING-WATER. 419 ess of filtration is at once laid bare. Bacteriological purity of well- waters can also be satisfactorily controlled," Professor W. H, Horrocks,^ too, remarking on the fact that, if a considerable time has elapsed since the occurrence of pollution, the bacteriological detection of the same, especially when waters of great original purity are concerned, becomes more and more difficult, adds : " It is, therefore, evident that a bacteriological examination has its limits of usefulness, and a slavish adherence to it under all conditions, combined with neglect of the hints to be obtained by chemical means, may lead to a perfectly erroneous judgment. Still, there is one branch of hygienic study in which bacteriology must always reign supreme ; it is now acknowledged on all sides that the working of sand filters for public water supplies cannot be properly kept under control except by appealing to bacteriological methods of examination." A positive result, the first instance in which the organism isolated responded to every test, including growth on gelatin, potato, litmus milk, bouillon and glucose bouillon, agglutination, and PfeiflPer's test with animals, is recorded by Drs. Kiibler and Neufeld.^ In this case, the cause of the disease lay in the use of water from a well infected by the urine of a person sick with the disease. Four weeks from the time of the first examination when the bacillus was isolated, a second analysis was made, which yielded bacilli which responded to all the tests excepting Pfeiifer's, which exception was supposedly due to modi- fied virulence. No colon bacilli were present either time. A second instance is recorded by Fischer and Flatau,^ who isolated the organism from a well-water in Pellingen. Similarly, a second attempt, made four weeks later, was unsuccessful. From what has gone before, it may be said that neither chemical nor bacteriological analysis is infallible. Each has its uses, and each may be helped by the other. The value of either lies in the skill displayed in interpreting the results, and this requires as much knowledge as the making of the examination itself. 1 Bacteriological Examination of Water, London, 1901, p. 3 ^ Zeitschrift fiir Hygiene und Infectionskrankheiten, XXXI. (1899), p. 133. ' Centralblatt fiir IBakteriologie, etc., XXIX. p. 329. CHAPTEE V. HABITATIOXS, SCHOOLS, ETC. Section 1. GENERAL CONSIDERATIONS. It is essential to hejilth that the liouses in Avhieh we dwell shall be built upon proper sites, free from dampness and organic pollution ; they should be provided with adequate means for heating, ventilating, and lighting ; they should be well supplied Avith Mater for general domestic jjurposes, and jjrovided with a proper system of plumbing for the removal of sewage ; they should be constructed with proper precautions against dampness from without or below. Heathu/, venti- lation, lighting, and plumbing are considered below under their several headings. Aspect. — It is commonly directed that hal)itations should Ik- placed so as to face the south ; but, unfortunately, one is not always in a posi- tion to be over-particular in the matter of points of the compass, and, indeed, there seems to be no particularly good reason why that side of the house in which is located the main entrance should face south and the others resj^ectively north, east, and west. The southerly side of a hill is very much to be preferred to the northerly, because of the greater amount of sunlight and of protection against the cold winds from the north ; but in a plain and elsewhere, whichever wall of the house ftices south, there must be, if the structure be rectangular, one to the north. Far better is a location with the corners of the house pointing north and south, for in that case every window must receive some direct sunlight at some ]iart of the day, whereas with sides facing directly north and south, the m indows of the former receive no direct sunlight and the rooms are dull and cheerless. In large cities, aspect is commonly a minor consideration, the desirability of a house being determined mainly by other circumstances. In general, it may be said that, when possible, a house should be so situated as to insure an abundance of light and air with protection against the cold winds of Avinter. Construction and Arrangement. — Consideration of building ma- terials and the details of arrangement of rooms and division of floor- space are l)eyond the scojie of a work of this nature, and it is necessary only to call attention to the importance of insuring dryness, light, and air, and such thoroughness of construction as shall not permit a too generous amount of natural ventilation with consequent waste of heat. Of the very greatest importance is the character of the cellar, that ]iart of the house which is most neglected during both construction and occupancy. Unless the site is one of unusual dryness, the cellar flroving its construc- tion. Desks and chairs should not be supplied with reference to age, but according to the size of children, and should be adjustable to each child. Common faults of chairs include improper shape of back, improper height, too great dej>th or breadth of seat, too much slope from front to back^ and improper horizontal distance from the desk. Common faults of desks include insufficient or too great height, and improper slope of the surface. Chairs should l)e of such a height that a\ hen the leg and thigh form a right angle the foot shall be squarely on the floor. If too high, the child cannot touch the floor, and fails to obtain the needed assistance from the feet and legs for the maintenance of an upright position ; if too low, the position is cramped and awkward and the child is forced to extend tlie legs in one direction or another, and to contort the body accordingly. The seat should be sufficiently wide to supjiort both thighs comfortably, but not so wide as to permit the assumption of bad postures in which the back is not well supported. It should not be too deep from front to back, since then ]>ro])cr attitude is imjiossible. It should slope very slightly from front to back, or be made slightly concave, in order that the tendency to slip forward may be counter- acted. The back should be curved forward, so as to support the child's back, and to be comfortable in whatever legitimate attitude he may assume, for he recpiires changes in ])osition for the relief of downward pressure and strain of muscle and ligaments, since any attitude long maintained results in fatigue. AVith chairs of the best form of con- struction, faulty attitudes are less comfortable than proper ones. Desks should be of such a height that the forearms of the child may be resti'd without causing, on the one hand, stoo])ing, or, on the other, raising of the shoulder and curving the spine. If too high for the child, the work is brought too near, and causes straining of tlie eyes. VENTILATION AND HEATING. 423 The top should have a proper slope downward of from ten to twenty degrees, and its edge should project slightly over the forward edge of the chair, so that the body and head may not be inclined forward too far. Proper height of seats and desks and correct horizontal distance between the two are attained by the adoption of adjustable furniture, of which there are many varieties. Blackboards should be dull black, and never shiny, for if they are shiny they reflect light, and what is written thereon is difficult to read from certain points in the room. They should be kept well cleaned, in order that the contrast with the chalk shall be sharp. The chalk should be white or yellow ; blue, green, and red chalk marks are much more difficult to read. Blackboards should never be placed between windows, on account of glare. Copying from blackboards is very try- ing to the eyes, on account of the constant necessary change of focus. Legislation Concerning Schools. — Many of our States have enacted laws providing for school sanitation in several particulars. Seventeen require expert examination of plans for projected buildings. Sixteen provide for fire-escapes and other protection from fire. A number provide for proper ventilation, but only two, Massachusetts and Con- necticut, have established a standard and requu'e its enforcement. Kentucky alone provides standards for lighting, for floor space and air space, for proper seating, and for water supply. About one-third of the States and Territories compel vaccination, and one-fourth provide safeguards in the matter of contagious diseases in general. Section 2. VENTILATION AND HEATING. We have seen how necessary it is to life that the CO2 given oflP by the blood in the lungs to the inspired air shall be discharged continu- ously from the body, and we know that whatever other effects the im- purities of vitiated air may produce, the effect of undue COj in the air is to interfere with the function of respiration. Therefore, it follows that the air which we breathe should be as free as possible from the impurities which we continually discharge into it, and that this condi- tion can be obtained only by constant dilution of them by a constant supply of fresh air. In the open air, this dilution goes on without artificial assistance and requires no consideration ; but in confined spaces, we have to a certain extent a reproduction of the conditions that obtain in the lung ; namely, the presence of a volume of vitiated air, separated from the purer surrounding air, and requiring to be dis- charged and replaced. In other words, the air in the first confined space, the lung, is discharged into the second confined space, the room, and thence must be removed to the outside and replaced by an equal amount of normal air. The constant dilution and removal of impuri- ties due to the necessities of life, so that their amount shall be so small as to be harmless, are the function of ventilation, which may be re- garded as the respiration of a building. 424 HABITATIOyS, SCHOOLS, ETC. It is, of course, not to be expected that the air of an inhabited room can under the usual conditions be maintained in a state of purity like that of the outdoor air, even though but one candle or one person be present to exchange carbon dioxide and Avater ft)r oxygen, but the impurities can be reduced to a minimum by the introduction of a proper amount of fresh air. AVhat shall be considered a proper amount, of continuous air supply depends upon what we adopt as a limit of per- missible impurity, measured by the amount of CO2 present. For the maintenance of a fair degree of vigor and stability through proper oxidation of the blood and dilution of the effete matters dis- charged, and for the maintenance of the fullest and most perfect func- tional activity, one requires respectively about 30 and 50 cubic feet of air per minute. Less than 30 will inevitably produce impaired vitality ; more than 50 can be productive of no gain in improvement, so far as the effects of the ordinaiy vitiating matters are concerned. At the latter rate, then, an hourly supply of 3,000 cubic feet is necessary for the proper dilution of the respired air of each individual present in a confined space. Thus, a room of 3,000 cubic feet capacity', inhabited by one person, should receive its full capacity of fresh air once every hour. But this renewal should be a continuous process, so as to prevent the accumulation of impurities which would occur if the air were replaced simply in bulk by an hourly aeration by opening win,- dows for the requisite few minutes. Xor should it be supposed that even with constant fresh supply the air of the room can have the same composition as that wliich enters from without, for the impurities of each respiration are not removed in separate, distinct lots, but are mingled in the general bulk. If the occupant's head were in a conduit l)riiiging the constant supply of fresh air and carrying away the j^roducts of respiration, no such amount of air would be necessary, and no contamination of the general su|)p]y would occur. Under ordinary circumstances, witli an hourly su])ply of 3,000 cubic feet per capita, the amount of CO, will not range above 6 or 7 volumes in 10,000, and any system of ventilation that will keep it down to this may be called good. Other impurities tlian those of respiration are to be considered in all questions of ventilation. The influence of burning illuminating mate- rial on the composition of air is very great, both as to the consumption of oxygen and the production of CO., and other impurities, and it is not insignificant in its relation to the temperature. The impurities from 1 cubic foot of ordinary illuminating gtis are such in amount as to require, according to various estimates, from 500 to 1,800 cubic feet of air foi- their proper dilution. Tluy include not only carl)on dioxide and water, but carbon monoxide, sulphur acids, nitrogen acids, marsh gas, ammonia compounds, unconsumed carbon, and other matters. Different forms of burners consume different amounts of gas to pro- duce the same illumination ; ordinarily from 3 to 6 cubic feet are used per hour, requiring 1,500 to 10,000 cubic feet of air-supply for proper dilution of the impurities produced by each burner. Therefoi'e, on AMOUNT OF SPACE REQUIRED FOR GOOD VENTILATION. 425 both hygienic and economic grounds, the burner which produces the maximum of light from the minimum of gas is the best for use, it being understood that a given volume of gas will yield the same amount of impurities, whatever the burner employed. The impurities from candles and lamps are less in number and variety, but, measured by their comparative illuminating power, they are larger in amount than from gas. For example, the combustion of an amount of candle or kerosene oil necessary to produce the same intensity of light as 1 cubic foot of gas will produce from 40 to 160 per cent, more impurities, and requires, therefore, proportionately more air for their proper dilution. The subject of ventilation, involving, as it does, the continuous introduction of pure air to displace that which has become vitiated by whatever cause or heated to such a degree as to be inconsistent with comfort ; having to deal with buildings and rooms of various sizes, designed for different uses ; and, as it is chiefly in the colder months that its importance is greatest, being intimately connected with the problems and cost of heating, is a very complex one which will not permit the adoption of inflexible rules applicable to all cases. Amount of Space Required for Good Ventilation. If it be agreed that for the most perfect results an adult requires an hourly supply of 3,000 cubic feet for the removal of his own effete matters, to say nothing of the accessory impurity dependent upon illumination, the next question is as to the amount of cubic space necessary per capita, that is to say, through how small a space that amount of air can be drawn hourly without disagreeably perceptible draughts. The sensation of draught is governed very largely by the temperature of the moving air, a cold slowly moving current being more perceptible than a warm one moving at a greater rate of speed. Draughts ^vhich are productive of discomfort are more dangerous than the ordinary vitiation of the air, and, therefore, complete ventilation with draughts is worse than partial ventilation without draughts. It has been shown by Petteukofer that, with the aid of delicate apparatus and mechanical power, about 2,500 cubic feet can be passed without draughts through a space of 424 cubic feet in an hour ; that is to say, through a room 8 feet square and 6.5 high, the air can be renewed six times. The minimum space 'svithin which one can receive, under artificial conditions, six complete changes of air in one hour is, therefore, 424 cubic feet ; but in order to get 3,000 cubic feet in one hour, the air would have to be changed seven times, and so the required minimum space would be 500 cubic feet. In large spaces, however, it is possible to obtain more frequent changes without danger from draughts. Thus, in a hall 40 by 20 by 1 5 feet, 40 persons may be supplied with 3,000 cubic feet each per hour, and each one will have 300 cubic feet of air 426 HABITATIONS, SCHOOLS, ETC. space. Therefore, in dealing with large spaces, we may assume 300 cubic feet per capita rather than 500. In the ventilation of small spaces, there is, in addition to the possi- bility and danger of draughts, the grave difficulty that the inlets and outlets are necessarily so near together that much of the air will pass directly from the one to the other without having mingled Avith the main body of air, and without, therefore, doing the slightest service in diluting the impurities present. In large air spaces, this objection does not apply with equal force, for the ttpportunity for diifusiou is greater, the larger the space, although, of course, here, too, the inlets and out- lets may be so placed as to favor the formation of direct currents. In the ventilation of large spaces in -which large numbers of people gather, as churches, theatres, schools, and lecture-rooms, we are at once coni'ronted by the fact that 300 cubic feet of space is a more liberal per capita allowance than is often practicable, and that this space is incom- patible with a draughtless ventilation by the necessary air volume. If, then, the question be asked, how to provide the necessary amount, there is but one ansAver ; namely, that it cannot be done. Fortunately, however, the danger from impure air is proportionate to the length of time of exposure, an occasional short time spent in a crowded room or public conveyance filled with bad air being less harmful than prolonged and habitual occupancy of a room in which the air is less vitiated, but yet not good. In general, it may be said that the importance of ventilation varies with the particular air spaces ; those which are used only at intervals and for short periods have much less need of it than those which are used uninterruptedly ; those which are not crowded demand less than those that are ; those used only for the moment need no consid- eration whatever, the natural forces at work at all times being suffi- cient for their needs. To insist upon thorough ventilation of every part of a house at all times, as most amateur hygienists do, is to demaud a needless waste of energy and money, for exccjit in the warm months when the windows and doors are left freely open for the sake not alone of ventilation, but of general comfort, ventilation goes hand in hand with heating and divides the expense. Nor can one successfully insist upon any rule that each person must have at least 1,000 cubic feet of air space with renewal of the contained air once in twenty minutes, for to do so is to urge in the case of the poor of large cities an impossibility, since space is costly, and, with ventilation to the proper extent, is beyond their means. Natural Forces in Ventilation. Before proceeding to the subject of systems of ventilation, we must consider the natural forces which are at work in the presence or absence of all schemes and systems. These forces are diffusion and gravity. NATURAL FORCES IN VENTILATION. 427 Diffusion and Gravity. — The rate at which gases diffuse is by no means the same for all, the lighter diffusing much more rapidly than the heavier. In fact, the rate varies inversely as the square roots of their densities. The province of diffusion in ventilation is limited to bringing the air to a condition of more or less complete homogeneity by causing the gaseous matters to become distributed throug-hout the mass ; but in an inhabited room it can do but little toward keeping the air at its normal composition. By reason of the law governing the rate of diffusion, there must of necessity be a constant, though slow, removal of gaseous impurity into the external air, for wherever two gases are brought into contact, diffusion will occiu', whether the meeting place be large spaces, as rooms, or small spaces, as pores in the plastering, bricks, mortar, stone or other material which forms the boundaries of that room. This force is, however, very inadequate, and can be of service only as an assist- ant to another. Moreover, it can affect only gaseous and not sus- pended matters. Of vast importance, however, is the other force, that of gravity. Equal voliunes of air at the same temperature and under the same press- ure will have the same specific gravity ; if the temperature of one of them be raised, it expands a definite amount for each degree, and thus has less specific gravity than the other the more it is heated. Being surrounded by air which is heavier than itself, it rises, or, more properly, is forced upward by the heavier air, which descends to occupy its place in the same way that a volume of light oil in a cylinder is forced upward when water is poured upon it. If, on the other hand, it is cooled, its volume contracts, its specific gravity is increased, and it sinks downward through the warmer lighter air below it. In this way, differences in temperature cause constant movements in bodies of air, and currents are established. In an inhabited room this force is always at work, for there must necessarily be some source of heat, even though it be only the body of the occupant. The air in contact with the body becomes heated, and is then displaced by the colder air about it; this in its turn is subjected to the same influences, and the whole of the contained air rises in temperature and is correspondingly ex- panded. As it becomes lighter than the surrounding air, the latter forces itself in and the original air out through all the available openings, and thus a certain amount of ventilation is accom- plished. Under the ordinary conditions of occupancy of a house or room, we have additional sources of heat in the combustion of fuel and illumi- nating materials, and no matter how imperfect the applied system of ventilation may be, and in spite of all efforts to exclude the external air, a very considerable amount of interchange of air is inevitable. It is only in a chamber that is to all intents and purposes hermetically sealed that no ventilation will occur when there is a difference be- tween the internal and external temperatures. Heated au' will escape through flues, through cracks around windows and doors, between the 428 HABITATIONS, SCHOOLS, ETC. boards of the floors and of the general structure, and even through the interstices of unpainted plastering and mortar, and through the pores of bricks. That a large volume of air Mill pass through cracks, needs no demonstration ; the passage of air through bricks, plaster, and mor- tar may easily be shown. If to the opposite sides of a brick, we fasten by means of sealing wax two ordinary glass funnels, and then smear its entire exposed surface with a liberal coating of the same material, all of the external pores excepting those within the spaces covered by the funnels are made impervious to air. If now we connect by means of a rubber tube the funnel on one side with a bottle in which air can be compressed by means of water pressure, and by the same means the Fig. 38. V Apparatus for demonstrating the permeability of bricks, etc., to air. other funnel with an inverted test-tube filled with water, and apply pressure, the passage of air through the pores of the brick will be manifested in a few minutes by the escape of bubbles from the outlet tube upward through tlie wtiter. (See Fig. 38.) The passage of air through plastered walls is much impeded by wall paper, and niay be totally prevented by oil paint and moisture. Numerous experiments have proved that with varying differences between the internal and external temperatures, the air of a room or building will be renewed ]xirtly, wholly, or repeatedly every hour^ even when eftbrts are made to prevent as far as possible the entrance NATURAL FORCES IN VENTILATION 429 of the outside air. The results vary with the difference in conditions, the highest effects being produced when the temperature differences are wide, the opportunities for leakage great, and the external air in active motion. With perfect calm and equal temperature, the result will be nil. Perflation and Aspiration. — Inequalities in outside temperatures give rise to the larger currents of air which we know as winds. These have a very great influence on ventilation both by their perflating action and by aspiration. The highest results of perflation are those obtained when obstacles to the free admission and exit of wind are removed, as, for instance, by opening windows in its path. The quickness and fre- quency of renewal of contained air by this means will necessarily depend Fig. 39. Fig. 40. Fig. 41. Common forms of stationary ventilating cowls. Rotary cowl. upon the size of the openings and the velocity and direction of the wind. The least effects are produced, whatever the velocity and direction, when the obstacles to entrance are greatest. The aspirating influence of wind is shown by the upward currents produced in flues when the internal and external temperatures are equal. A current of air, moving swiftly across the outlet of a flue, has the same effect on the contents of the flue as that of a common hand atomizer has on the contents of the tube. The air in the upper part is carried along mechanically, a partial vacuum is formed, and that which is below rises to take its place, and is in turn carried away. In the case of the flue, air from below is constantly drawn up and dissi- j)ated ; in the case of the atomizer, the liquid into which the tube dips 430 HABITATIONS, SCHOOLS, ETC. is lifted and blown into spray. This influence is utilized and assisted by various forms of coavIs placed over outlet flues ; some of these, however, although they seem to be an aid, are really a hindrance to the outflow, as may easily be demonstrated. In Figs. 39 and 40, are shown forms of cowls which offer some assistance to the aspiratory influence of winds, and in Fig. 41 is shown another very popular kind, the rotary cowl, which offers an obstruction to the passage of air. As the wind causes the top to revolve, the im- pression is made that work is being performed ; that in its revolutions it is creating a suction which causes an upward current of air. As a matter of fact, however, it is doing no such work, but is, on the con- trary, interposing an obstacle to the passage of air. This may easily be demonstrated by measuring by means of an anemometer the amount Fig. 42. Aspirating cowl with vane. of air discharged through the flue during a given period while the re- volving top is in place, and again during an equal period while it is removed. The difference between the results obtained will invariably be in favor of the period during which the cowl is absent. Other forms of cowls, constructed so that their outlets arc turned by means of a vane away from the wind, are useful in assisting aspiration. Such a form is shown in Fig. 42. By reversing the position of tlie vane, the mouth of the cowl is turned toward the wind so that the flue is converted into an inlet for fresh air. NATURAL VENTILATION. 431 Natural Ventilation. Ventilation that proceeds from the operation of natural forces is known as " natural ventilation." For the attainment of the largest results, these forces must be assisted to the extent of removal of ob- stacles to their action so far as may be advisable. It is not well to depend upon the chance cracks and upon the migration of air through the pores of building materials, but necessary openings, both inlets and outlets, should be provided. The greater the obstacles to the escape of heated air, the less the opportunity for successful natural ventilation. The extreme of obstruction to the escape of contained air may be illustrated by an hermetically sealed metallic box or by a closed glass bottle. Suppose we provide one small opening in the side of the box or in the stopper of the bottle to act as an outlet and inlet, and observe the result. According as the contained air is warmed or cooled, the opening will act as an outlet or as an inlet, but only to a limited extent. The expansion due to heating will cause the escape of a portion of the contents ; the contraction due to cooling will cause the indrawing of some of the outer air ; but in either case, the movement is all one way, and there is no real interchange. Suppose, however, two openings are supplied ; then one may act as an outlet, the other as an inlet, and a constant inward and outward current may be main- tained. The more tightly fitting we make our windows and doors, and the mere impervious to air we make our walls by means of paint and sheathing paper, the more do we oppose natural ventilation. On the other hand, the intelligent placing of inlets and outlets furthers the object to be achieved. In addition to permanently installed inlet and outlet flues, tempo- rary openings may be utilized whenever desirable. The most avail- able of these is the opened window, which may be utilized so as to avoid too voluminous exchange and unbearable draughts. The area of the opening may be very simply regulated, and the air may be deflected upward or the current may be broken up by the interposition of fine wire gauze, flannel, or other pervious material. A very common plan is to place a board lengthwise under the lower sash, so as to fill completely the opening made by raising the window, and thus establish an inlet or outlet where the sashes overlap each other, for the barrier to the movement of air, formed by the juxtaposi- tion of the lower border of the upper sash and the upper border of the lower one, no longer exists, and the entering current, moreover, is given an upward direction. Instead of a board, a frame, over which a diaphragm of flannel is fastened, may be used. This arrangement is pervious to air but impervious to dirt, which, therefore, is filtered out. Movable panes, either sliding or swinging by the side or end, are frequently employed, especially ' in double windows. There are also numerous patented devices for window ventilation, all designed with the idea of dividing or deflecting the current of admitted air. 432 HABITATIOyS, SCHOOLS, ETC. As lias been remarked above, the most important force in natural ventilation is that dependent upon unequal tem])ei'atures of bodies of air ; in a perfect calm and with equal temperatures, natural ventilation would have to depend wholly on the force of diffusion. The enormous influence exerted by the heating and lighting of a building or room on its ventilation becomes, then, self-evident, but it is not simply as a motive force that the relation between heat and ven- tilation is so close and important, for the incoming air must be raised to an agreeable temperature in order that the space may be habitable. Thus, a very large share of the cost of heating is chargeable to venti- lation, whatever the system of ventilation be. In the matter of expense, the amount of leakage through cracks and other small open- ings is, in a certain class of cases, of very great importance. In our dwellings, it is important that the interchange of air shall proceed continuously in the inhabited parts, but in buildings which are used only by day, and perhaps for only a few hours daily (schools, etc.), it is not essential that the air shall be renewed constantly at other times ; and here it is Avise to obstruct the leakage as nmch as possible by perfect construction and by dampers in the flues, so that waste of heat, fuel, and money may be prevented. For the })romotiou of the process of natural ventilation, a number of " systems " have been devised, many of which can be ])roductive of no results other than incomes for their promoters. As a rule, most of those noticed in works of this character are either ill-adapted to the conditions of our climate or inconqiatible with our methods of building, and mIII, therefore, not be considered here. The only system of natural ventilation Morthy of advocacy is that which provides })roper inlets and outlets and a suitable means of heating. Inlets and Outlets. — As to the size, location, and number of in- lets and outlets, no hard-and-fast rules can be ap})li('(l for all cases, since the conditions are widely varying, and many diflerent circum- stances have to be taken into account. But general rules may be laid down. If several inlets are to lie })rovided in a room, it is essential that they should be distributed in such a manner as to insure a thorough blending of the admitted air. They should not be so placed, with reference to outlets, as to favor the forming of direct currents between them, whereby a large proportion of the infloA^ing air is discharged without having fulfilled its function — a not unusual condition, which illustrates that the amount of air admitted is not by any means a measure of thoroughness of ventilation. Their location is not such an important matter as the placing of the outlets, but, in general, an inlet is placed best on an inner wall where it shall be most nearly central in relation to the outside walls. A\'ith referenci' to the floor, if the incoming air is heated, inlets may l)e placed high or hnv ; but if it is admitted cold, they should be at a higher level than the heads of the occupants, and ])rovided NATURAL VENTILATION. 433 with arrangements for deflecting the current toward the ceiling. This may be accomplished by causing the current to impinge upon a surface slanting upward. The results of this deflection are that the fresh air becomes mixed with the warmer air, and that more time is required for it to reach the lower parts of the room, when it will have become sufficiently warmed not to cause discomfort. The interposi- tion of the deflecting surface also spreads the current radially and reduces its velocity. The incoming air becomes mingled with the general supply and joins the currents which are constantly in motion. That which comes in contact with cooling surfaces, such as windows and outside walls, is cooled, and, therefore, falls toward the floor, and that which takes its place as it falls is cooled in its turn and fol- lows after, so that currents are established, which tend to keep the whole bulk in more or less rapid motion. As these currents reach the floor, their natural trend is across that surface toward the inner warmer walls, where they become heated and are inclined toward the ceiling, reaching which, they are pushed by the force behind and drawn by the one in front toward the outer walls and w'indow^s again. In the meantime, some of the air is escaping through outlets, and diffusion of the impurities is proceeding, so that a more or less even character is brought about throughout the air of the room. Outlets may be placed at the level of the floor, in the ceiling, or at any height in the walls, according to the conditions of each individual case. If the incoming air is not heated, the outlets should be placed high up, for where only unheated air is admitted, the warmest air must be the oldest and its location will be in the upper air space. If, on the other hand, the air is heated, the outlets may be anywhere so far as height is concerned, but there is some choice in locations with respect to inner and outer walls. Outlets placed beneath windows or near outer walls will \vithdraw the falling currents of the only recently in- troduced air before it has had an opportunity to become well mixed by passage across the floor to the other side, and before it has in any proper degree fulfilled its functions ; but if its passage through the lower strata is not interrupted in this manner, it is enabled to mix with and dilute the impurities of the air already vitiated, and thus effect a large measure of work, and so when it reaches the other (inner) side and finds an outlet for its escape, there is no objection to its withdrawal, and, indeed, its removal then is highly desirable. Hence, and for another reason as will appear, outlets should be placed in inner walls rather than near or in outer cooler ones, and near the floor where they may intercept the air before it may again become a part of the ceiling currents. If but one outlet is to be provided, it should be placed with reference to the most even movement of the current over the whole area, having in mind the fact that the air movement toward it is convergent, and the direct reverse of the flow from the deflecting and diffusing surface GO at the inlet. As to size, it may be said, in general, that a single outlet, or the 28 434 HABITATIONS, SCHOOLS, ETC. aggregate if there be more than one, should be of such size as to insure the possibility of removal of such an hourly air sui)ply as the space is likely to require under the ordinary conditions of its usual occupancy ; that it should not materially exceed this limit ; and that the final velocity of the outflowing current should not be productive of the sensation of disagreeable draughtiness. As to the shape of inlets and outlets, it is self-evident that, with equal areas, that which has the smallest periphery will oifer the least frictional resistance, and is, therefore, best adapted. Thus, a circle en- closing an area equal to a square foot has a smaller periphery than a square enclosing the same area, and a square has a smaller one than an oblong rectangle. Take, for instance, a square foot ; it may be in- cluded within boundaries : 12 X 12 inches, a square. 16 X 9 18 X 8 24 X 6 36 X 4 • oblong rectangles 48 X 3 72 X 2 44 X 1 , With these boundaries, the periphery ranges from 4 feet (the smallest) to 24 feet 2 inches. The frictional resistance will, therefore, be greater in proportion as the shape varies from the circle and square. The shafts communicating with the inlets and outlets should be so disposed in the general plan as to oifer the least resistance to the inflow and outflow of air. Unless they are heated artificially, inlet flues should not be located in outer walls, on account of the likelihood of the formation of down draughts due to cooling of the air column. Their inner surface should be as smooth as possible, in order to bring to a minimum the loss of movement due to friction, and they should be cylindrical, if possible, for the same reason. They should be as free as possible from angles, and especially right angles, because of the very serious loss of motion which these cause, each right angle dimin- ishing the current about half; thus, after passing one right angle, the flow would be half; after a second angle, the half would be reduced to a quarter, and after a third, to an eiglith. The neglect to take into account the loss of flow by friction, bends, and angles is responsible for the failure of many a plan for ventilation. Wiiat has been said concerning the impossibility of making general rules for the sizes of inlets and outlets applies with equal force to the fixing of sizes of flues, for these depend upon the many and varied conditions which, even under the best favoring circumstances, affect the rate of flow. In planning inlet and outlet shafts, it is to be borne in mind that something more is necessary for their proper working than the dictum that this one is for fresh air and that one for foul, for natural forces have no respect for mere names and plans, and the current in a flue will be upward or downward, inward or outward according to natural ARTIFICIAL HEATING IN ITS RELATIONS TO VENTILATION. 435 laws. Outlet shafts may become considerably cooled by low external temperatures ; they may be invaded by rain and snow, the evaporation of which causes cooling and, therefore, increase in gravity ; or there may be an insufficiency of inlet air, so that a partial vacuum is formed by the current of one large outlet flue, which thereby causes a reversal of that of a smaller one, so that one flue draws against another. It is from any one of these causes that a chimney may fail to discharge smoke upward — a circumstance noticed more often in summer than in winter. Mechanical Ventilation. Mechanical ventilation consists in the propulsion or extraction of air by means of blowers or exhaust fans driven by steam or electricity. That in which the air is propelled by the action of a blower is known as the " plenum " system, and the other, in which it is withdrawn by am exhaust fan, is known as the " vacuum " system. In the plenum system, the air is drawn into a box, in which the revolving blades of a fen are located, and it is then driven into a cen- tral conduit, and from there through appropriate shafts to the spaces for which it is intended. When it is desired, the air may be received from or blown first into a chamber where it may be heated. The air supply may be regulated very easily by diminishing or increasing the number of revolutions per minute, but it should always be in slight excess of the real need, in order to produce a slight outward pressure, which will prevent inward leakage. In the vacuum system, the air is extracted from the various rooms through pipes leading to a central shaft, where it is drawn into the fan. and discharged outwardly. This system has among other disadvan- tages that of great inequalities in draught in the different discharge tubes, and that the vacuum condition favors the inward leakage of cold air through cracks, walls, and about windows, and tends to cause cold floors and disagreeable small draughts about windows. In consequence, more fuel is needed for the maintenance of a proper temperature, and the system is, therefore, a source of greater expense. The advantages of mechanical ventilation lie in the fact that the object sought is attainable in any and all conditions and variations of weather, and that less space is required for shafts than in the case of natural methods. Mechanical ventilation on a comparati\^ely small scale is employed commonly in crowded offices and other spaces by means of small ex- tracting fans run by the aid of electricity (connection being made with the electric light system) in specially provided locations connected with outlet flues, or directly in a space made by removing window panes. Artificial Heating in Its Relations to Ventilation. First, for the proper understanding of the subject of heating in its bearings on ventilation, it is necessary to consider the different ways 436 HABITATIONS, SCHOOLS, ETC. in which heat is imparted. These are three in numl)er : radiation, conduction, and convection. Radiation. — Radiant heat passes from its source through the air to bodies by which it may be absorbed, transmitted, or reflected. Air, being " transparent " to heat, is not materially aifected, and the drier the air, the less heat it will retain. It passes directly from its source in waves, like the waves of light, to the object upon which it falls, and the amount reflected or absorbed varies with the nature of the object, its color, the character of its surface, and its temperature. Its in- tensity varies inversely as the square of the distance between the source and the object upon which it falls ; thus, the amount received by two objects 1 and 5 feet respectively distant, will be inversely as 1 and 25 ; at 1 and 10 feet, inversely as 1 and 100 ; at 5 and 10 feet, inversely as 25 and 100 ; that is, the nearer will receive in the first in- stance 25 times ; in the second, 100 times ; and in the third, 4 times as much as tlie farther object. As an instance of radiant heat, we may take that which proceeds from an ojjen fire. The heat passes in direct lines through the air to the walls, floor, ceiling, furniture, and other objects in its path, and these absorb some and reflect the rest to other parts of the room. It directly warms only that surface of an object that is directly opposed to it. The ol)jects by which it is opposed then disseminate it in two w'ays : by conduction and convection. Conduction. — Conducted heat is that which passes from one par- ticle of matter to another in direct contact ; that is, from one particle to another of the same object, or from one object to another which it touches. Conduction acts through all solid substances, but by no means to the same extent, some being good, some indifferent, and others bad conductors. The best conductors arc metals, and these vary within very wide limits ; copper, for instance, is a very much better conductor than iron or zinc. Wood is a poor conductor, and woven and felted materials and asbestos are very jioor. Through liquids and gases, heat is conducted to only a very limited extent, but there is no substance known that is absolutely non-conducting. Good conductors permit a rapid flow of heat thi'ough their substance ; poor ones, only a slow transmission. Good conductors relinquish their heat rapidly to their colder surroundings, whether air or anything else, and withdraw heat from bodies which are warmer than themselves. Convection. — Convection is the ]irocess l)y which heat is conununi- cated to gases and liquids, acting through their mobility, which ])ermits those ]iarts that have been expanded by reason of becoming heated to be displaced upward by cooler portions, which, in their turn, receiving- heat, give way to others, until the whole mass becomes raised in tem- peratui'e bv continued application of heat and consequent maintenance of circulation. Every warm body with which air comes in contact communicates its heat to those portions in its immediate vicinity ; these expand and are forced onward by the cooler heavier parts nearest them ; these in their ARTIFICIAL HEATING IN ITS RELATIONS TO VENTILATION. 437 turn give way to others, and convection currents are established to such an extent that the air of a room takes on a very complicated state of activity. Convection currents are established by every person in a room so long as the temperature is below that of the body. They are estab- lished by the warmer walls, floor, furniture, hot-water pipes, steam radiators, close stoves, and other warm ol^jects, and in this way the air becomes heated. The air which enters rooms through shafts com- municating with the air chambers of fm*naces and " indirect radiation " apparatuses are convection currents in the largest sense. The direct ravs of the sun, passing through windows and absorbed by the floor, walls, and other objects which they strike, also cause upward con- vection currents. Methods of Warming. — The principal methods of heating houses and rooms are : 1. Open fires, 2. Stoves. 3. Furnaces. 4. Hot- water pipes. 5. Steam pipes. The method most applicable in any particular case will depend upon the size of the room and the number of rooms in the building. In general, it may be stated that the smaller the space, the more simple the method. For a single room, an open fire or a stove will be sufficient ; for a small house, stoves or a furnace ; for a large one, one or more furnaces or hot-water or steam apparatus ; and for large buildings — office buildings, for instance — " direct " or "indirect" steam. 1 . Open Fires. — Practically the whole of the heat supplied by an open fire is radiant. If the fuel is held in a grate, there is, of course, a cer- tain amount of conduction from the bars, and of convection currents in the air in its immediate vicinity. But this heat does not get out into the room, because it is immediately carried up the flue by the draught of the chimney. The radiant heat is absorbed, reflected, and distributed in the manner already described, but reaches directly only those surfaces which are opposed to its source — which accounts for the saying that, in a cold room with an open fire, " one side roasts while the other freezes." Only a small part of the total heat of the fuel consumed is available for heating, since most of it — about seven- eighths — is carried at once up the chimney. An open-fire stove, such as the old-fashioned " Franklin," which stands out in the room, and is connected with the flue by stove piping, yields a large amount of its heat, since the material of its construction is heated by conduction and then gives it off to the air by convection. Open flues cause the introduction and removal of large volumes of air, but these are by no means always well mixed with the whole mass of contained air. Nevertheless, a large measure of ventilation is ac- complished, a certain amount of heat, perhaps sufficient for immediate needs, is given off, and there is also an unmeasurable addition to the general cheerfulness. They may cause too much draught, and they are certainly not economical, but as accessories to other heating methods they may be most useflil. 2. Stoves. — Close stoves have more direct results in heating and less 438 HABITATIONS, SCHOOLS, ETC. in ventilating than the open fire, for more of the heat produced is available, and they discharge into the chimney only the air volumes that have passed through them. The materials used in their construc- tion, iron, soapstone, brick and fireclay, conduct the heat and give it off to the air with varying rapidity ; cast iron yields it about as rapidly as it is received, soapstone and brick give it off only gradually, but for a longer period. AVhen cast-iron becomes red hot, it may be decidedly objectionable for two reasons : first, that the organic dust particles in its immediate vicinity become charred and yield odors ; and second, that it absorbs and transmits considerable carbon monoxide from burning coal. Stoves may be so arranged as to act not alone as heaters, but as ventilating apparatuses, and this fact is of very great value in the case of small school buildings in country districts. The stove, standing out in the Fig. 43. ^-^M Jacketed ventilating stove. room, may be surrounded by a cylindrical jacket from the floor up- ward, leaving a sufficient air space between the two. Through the floor within the enclosure, is an opening into an air duct comnmnicat- ing with the outdoor air. The heat of the stove is comnuinicatcd to the air between the latter and the jacket and an upward current is formed, which draws upon the fresh-air conduit, so that a constant current of warmed pure air is thrown into the room. (See Fig. 43.) It goes without saying, that here, as elsewhere, the incoming air must be taken from points where its purity cannot be interfered with by local conditions. Gas stoves and oil stoves have the advantage over others that they are more prompt in results, more easily controlled, and more quickly put out of use. They have the disadvantage, however, tluit the j>rod- iicts of their fuel combustion arc discharged directly into the air of the room. In the case of the oil stove, this is not such a serious matter, ARTIFICIAL HEATING IN ITS RELATIONS TO VENTILATION. 439 since the perfect combustion of good oil results in carbon dioxide and water ; but with gas the products are more numerous and varied, and include some that are irritating and poisonous. With proper ventilation, however, in the case of both, no harm will be done. 3. Furnaces. — Hot-air furnaces are not only of very great importance as heaters, but of enormous influence in ventilation. In their use, the cold outdoor air is brought in by a conduit, the "cold- air box," to a chamber in the upper part of the furnace, above and surrounding the "dome," where it comes in contact with the very hot surface and is heated by convection. Thence it passes upward through separate tin tubes to the several places for its discharge. In a house which is unprovided with special inlet and outlet flues for ventilation — and most of our houses are so constructed — a furnace of ordinary heating capacity performs an amount of ventilating work quite sufficient for all needs, and for which it rarely receives credit. It discharges into the various rooms a constant supply of warmed fresh air. Where and how it all escapes is a matter of secondary interest and importance, for it gets out wherever it may find its way. 4. Hot-water Pipes. — Hot-water heating depends upon the circulation of water by convection currents through a system of pipes which may extend all through a large-sized building. The water is heated in a boiler below and passes through a main, leading from the upper part thereof. As one portion of water comes in contact with the heating surface and expands, it is moved along, and the circulation becomes established just as with air. The " main " gives oif branches where needed, and these at their extremities turn back and become " returns," which eventually connect with each other and form the " main return," which, conveying the cooled water, enters the boiler at its lowest point. The first part of this system may be compared with the arteries, and the " returns," Avith the veins of the body. Yents are provided for the escape of dissolved air liberated from the water, and " cut-oflPs " are inserted for the shutting out of any part of the system as desired. It is very necessary that air should not be allowed to accumulate in the pipes, since it will stop the flow. In low-pressure systems, a small cistern is provided to allow for the expansion of the water and to pre- vent its overflow. The hot- water system may be of either high or low pressure. With high pressure, the pipes are smaller and necessarily stronger, and the water is heated to a considerably higher temperature (300° F.), and hence circulates more rapidly. With the low-pressure system, the water does not go much, if any, above 212° F. With the hot-water system of heating, the air is heated mainly by convection, though from polished pipes a certain amount of radiation occurs. With high pressure, the air may easih^ be overheated. 5. Steam Pipes. — In steam heating, the system is very like that of hot-water heating, except that steam is the circulating medium instead of water. With steam, and, indeed, with hot water, heat may be dis- tributed by the " direct " or " indirect " methods. In the " direct " method, the pipes are distributed within the space to be heated, and the 440 HABITATIONS, SCHOOLS, ETC. air of each room is heated separately. lu the "indirect" method, the heating surfaces are all concentrated in the basement, and are enclosed in galvanized iron conduits, which receive and conduct the air just as in the case of the hot-air furnace. The two methods, it will be noticed, vary widely in the matter of assisting ventilation ; the direct brings in no air, but heats that which is at hand ; the indirect brings in large volumes of heated fresh air, and thus insures change of air. In conclusion, may be mentioned the considerable heating and circulating influence of burning illuminating gas. By means of suit- able outlets above the burners, gas may be made not only to discharge the products of its own combustion, but to send out large volumes of otherwise vitiated air as well. Xor is the heat of the sun so insignifi- cant that it may be passed by without notice in the planning of sys- tems of ventilation. Inasmuch as the difference in temperature of the outside air on the north and south sides of a house averages about 5, and may reach 10, degrees F., just that amount of advantage may be gained by taking the air for ventilation from the warmer side. In gravity ventilation, the inlets should be where they may face the pre- vailing winds. Regulation of Temperature. In carrying out any scheme of efficient ventilation, it is necessary to guard against overheating, which may not be noticed until it be- comes so marked that it cannot help attracting attention. When such is the case, the common practice is to cause a lowering of the tem- perature to the desired point as soon as possible by opening windows to admit the colder air. The consequence is the jiroduction of a distinctly cold atmosphere, more so than ordinarily is shown by the thermometer, which does not react very promptly to sudden changes. This produces chilly sensations which call for a return to the original condition. In the meantime, a lot of heat has been wasted and the foundation for a cold has, ])erhaps, been laid. If windows are left open in the upper stories, as often happens in overheated buildings, there are constant outflow and waste of heated air, with a corres]>ontling inflow of un- warmed air below, which requires the expenditure of additional fuel in orde'r that the lower stories shall be properly wanned. In overheated buildings, tliere is also the additional loss from outward leakage through all possible outlet channels. To prevent Avaste of heat in properly heated buildings, we have recourse to double glazing and double windows. Double glazing is accom})lished by fitting two ])anes into each space, instead of one, with a space of a (juarter or a half inch between them. By this meiins, the loss of heat occurring through ordinary windows is reduced about one-third, which means a saving of considerable fuel, since the loss of heat by conduction through glass windows is very considerable. Double windows are still more efficient as heat-savers. Here the outer window is made to fit as accurately and closely as possible by NECESSITY OF PROVIDING MOISTURE. 441 the use of listing, and we have between the two windows a fairly deep space filled with air, which is a very poor conductor of heat. It is on the same principle that we use loosely woven woollen goods and furs, which hold within their meshes and between the hairs a large amount of this poor conductor. The loss of heat through walls is lessened when a similar air space exists within them ; a solid wall will con- duct a very large amount of heat and waste it, while the same amount of building material, or considerably less, may be so disposed as to bring this loss down to a minimum. Loss of heat is caused also by dampness of walls, for a continual evaporation goes on from their surface, and this requires heat and pro- duces cooling. Every ounce of moisture so vaporized requires the consumption of extra fuel. Necessity of Providing Moisture. Concerning the need of insuring a normal amount of moisture in the air of heated buildings, there is more or less difference of opinion, but the weight of evidence from a medical standpoint and from our own sensations points to the advisability of introducing an amount of moisture sufficient to bring the relative humidity of the air to 50 or 55 per cent. The lower the temperature of a body of air, the less the amount of moisture it can hold, and what would be saturation at a low tempera- ture would be but a very low relative humidity at a high one. For instance, a volume of air at 0° F., containing its fullest possible amount of aqueous vapor, admitted to the cold-air box of a furnace and then heated to 85° F. before being conducted to the rooms of a house, will have at its new temperature but a very small relative humidity. It will be so much drier than any outside air, that that of the driest climate in the world will be moist in comparison. The great majority of U. S. Signal Service Stations have a mean relative humidity of 65 to 75 per cent. ; only twenty-four show below 60 or over 80, and the very lowest is in the hottest part of Arizona, where newspapers crack when handled, glued furniture falls apart, and the skin becomes hard and dry. At this place. Fort Yuma, the mean relative humidity is 35 per cent. When outdoor air is heated so as to maintain an even temperature of 70° F., but with no addition of watery vapor, its capacity for ab- sorbing moisture is very much increased, and it will take it up from all moist objects with which it comes in contact. It will take it from the skin, from the mucous membranes of the mouth, nose, and respir- atory tract ; from furniture made from wood which, in the process of kiln-drying, was never brought to such dryness ; from the leather bindings of books, causing them to crack and fall to pieces ; and from plants, which, in consequence, wither and die. It thus causes more or less dryness of the skin, irritation of the throat, and cough. It causes also need of a higher temperature to give the same sensation of warmth 442 HABITATIONS, SCHOOLS, ETC. and comfort than is the case with air containing a normal amount of moisture. It is on account of the disagreeable and destructive effects of extreme dryness that water-holders are attached to furnaces and stoves so as to give moisture to the heated air. But even when atten- tion is paid to keeping them full, which is not often, they are very inadequate for the ])urpose. Air at 25° F., saturated with moisture and then heated to 70° F,, would need more than 0.5 pint of water in every 1,000 cubic feet to give it a humidity of 65 per cent., and this is far in excess of the capacity of the ordinary waterpot of the furnace, as is seen when we reckon' what 0.5 pint per 1,000 cubic feet means in the course of a day. Moisture may be imparted to the air by exposing pans or porous vessels of water to the heated current, or by means of the " humidi- fier," which exposes to the air passing through the registers a surface of cotton wricking comnuniicating with the reservoir of water. (See Fig. 44.) With this device. Dr. H. J. Barnes, of Boston, reports Fig. 44. Humidifier. that he is able to keep his office at 53 per cent, relative humidity by evaporating an average of 4.5 quarts of water per day. At the same time, he finds a temperature of 65° to be perfectly comfortable where before he had required 70° or 71°. On a larger scale, water may be vaporized into the air in the form of steam from a boiler. In the building of the American Bell Tele- phone Co., in Boston, a building having a capacity of 450,000 cubic feet and a day-time population of more than 450 persons, the air, which is distributed by the mechanical system, is drawm into the building at the rate of" 26,000 cubic feet per minute, heated to about 100° F. in the stack room, and nioi.>^tencd so as to contain about 50 per cent, relative humidity. For the production of tliis condition, no less than 675 gallons of water in the form of steam are given to the air in ten hours, or about one and a half barrels per hour. Certain parts of the building which, before the adoption of this process, had DETERMINATION OF RATES OF VENTILATION. 443 teen heated with some difficulty, are now made more comfortable, and in the whole building 3 degrees less heat are required for the mainten- ance of an agreeable temperature. According to Mr. C. J. H. AVood- bury,^ under whose direction the plant was installed, "another feature indicating the greater comfort of the building was the absence in win- ter of the coughing by those employed there, a cough of the bronchial liiud or from the larynx, a cough which ends with a squeal, which is so prevalent in Xew England during the winter, especially in those employed in offices." Filtration of Air. — Here may be given an instance of the benefit derived from filtering; larg-e volumes of air introduced into a buildino- for purposes of ventilation. In the building above mentioned, the air is drawn into and through a system of large cotton bags 30 feet in length, in which all dirt and dust is retained. About a peck per month is separated in this way from the air, which is drawn not from the street level, but far above it. An analysis, chemical and micro- scopical, made in April, 1897, showed 22.67 per cent, of organic and 77.33 of inorganic matter. The material consisted of all manner of animal, mineral, and vegetable substances ordinarily present in the dust of large cities. Determination of Rates of Ventilation. The estimation of the amount of air entering and leaving a room through inlet and outlet flues is a very simple matter, but the results may not be accepted as an indication of the efficiency of ventilation, since it so often happens that much of the effluent air has failed to per- form its full duty in diluting the impurities arising from respiration and combustion. Xevertheless, such a determination may yield im- portant indications. In order to ascertain the volume of air passing through an opening, whether inlet or outlet, it is necessary to know the area of the opening and the velocity of tlie current. The former is easily calculated arith- metically ; the latter can be found only by the use of an anemometer, an instrument of very delicate construction, which registers the distance travelled by a current of air in any period during which it is ex- posed. A current of air, passing through an opening, has not the same velocity at all points of its cross-section. It moves in the same manner as a river — faster at its center, where it is least subject to the influence of friction. Therefore, the velocity should be taken at diflPerent points, and the mean of the results accepted as its true rate of movement. The anemometer is held for a given time, say half a minute, at a point at the periphery of the opening, and then moved along a short distance and held for an equal period, and so on, from point to point, until the whole area has fairly been traversed. The reading of the instrument is then noted, and the distance indicated is divided by the number of ^ Transactions of the Xew England Cotton Manufacturers' Association, Vol. 63. 444 HABITATIONS, SCHOOLS, ETC. points where stops have been made. The quotient equals the distance travelled by the whole current during the unit of time employed. It will be found most commonly that the movement at the periphery is very slow, and that, as the center is approached, the velocity becomes greater and greater, the maximum being attained at the center. Know- ing the average movement in feet or meters, the volume is calculated by multiplying this by the area in square feet or square meters, the product being the volume in cubic feet or cubic meters passing during the unit of time. From this result, the volume per hour is easily made known. Example. — The size of the opening is 2 by 3 feet ; the area is, therefore, 6 square feet. The anemometer, held at twenty-four points for fifteen seconds each, registers 228 feet. The mean of this is 9.5 feet, and the current is moving, therefore, at the rate of 38 feet per minute. The cross-section of the current being 6 square feet, the volume discharged in a minute equals 6 X 38, or 228 cubic feet, and, in an hour, 13,680 cubic feet. By determining the rate of discharge through all inlets and outlets in this manner, an idea is obtained of the amount of ventilation occur- ring through means provided, but, as has been stated, not of its effi- ciency. The sum of the inlet discharge \\'ill almost never agree with that of the outlet, since much air enters and leaves a room through other openings. Knowing the capacity of the room, we learn from the amount of inlet air the number of times the air of the room has been replaced. The full measure of ventilation and its efficiency may be determined very closely by methods originated by Petteukofer. One of these con- sists in first creating an unusual degree of impurity either through respiration of a large number of persons, as, for instance, by children occupying a schoolroom, or by burning a number of candles, or by other chemical processes, then, after taking a specimen of the air for analysis, keeping the room closed for an hour or two. At the expira- tion of the allotted time, a second sample is taken, and from the results of the two analyses, the rate of ventilation is ascertained by means of Seidel's formula, which is as follows : C=^ 2.303 m. logfi^^-^- in which C ^= amount of air which has entered. 2.303 is a constant. m = capacity of the room. Pj ^ amount of CO,, ori.m i f iii I ' 'i'. fioom uith ordinarw \ Action of prismatic glass in projecting light. Fig. 46. tipped up and projected toward the opposite sides of the room, as shown in Fig. 45. Vertical section of a sheet of the glass is shown at A in Fig. 46. By varying the angle of the prisms, the conditions obtaining in any situation can be met and light may be projected in any desired direc- tion. Naturally, the prisms cannot be used indis- criminately, for a series adapted to light the entire lower part of the room with a certain sky angle might, when ap})lied to another, throw the light toward the ceiling instead of to the parts where it is required. Therefore, to meet all conditions, the glass is made with a great range of angles, and the particular kind needed in any situation is determined by measurement. Where the sky angle is very small, canopies, hung at the proper angle above the windows, serve to throw inward a flood of light. The disadvantage of prismatic glass is its great cost. Ribbed glass is very efficient and much less expensive. This is made with 4, 5, 7, 11, and 12 ribs to the inch, and of different thickness and weight, since the few^er the ribs, the deeper they must be cut, and the thicker, therefore, the A B Vertical section of pris- matic and ribbed glass. LIGHTING. 447 glass. Vertical section of a sheet of ribbed glass is sho^vn at B in Fig. 46. Artificial Lighting. — The methods of artificial illumination com- prise electric lighting and those de2:)endent upon the combustion of oils, gases, and hard fats. The oils employed are chiefly of mineral origin, but animal and vegetable oils are used to some extent, although not very much in this country. Hard fats in the form of candles are used very extensively in all countries, on account of safety, cheapness, and general availability. The gases in common use are derived from coal and hydrocarbons. Of late, acetylene gas, obtained by the action of moisture on calcium carbide, has come into extensive use. Luminosity of Flame. — In the combustion of a candle, it will be ob- served that the flame consists of four parts, the lowest of which, blue in color, gives out practically no light ; the middle portion, dark in color, consists of hydrocarbon gas generated from the substance of the candle ; next is the luminous yellow portion ; and outside of this, is an almost invisible envelope. The atmospheric oxygen, moving toward the inner portion of the flame, unites with the carbon escaping outward from the luminous portion, and forms carbon dioxide ; more oxygen passes onward and inward, meets the hot gas from the central part of the flame, and, being insufiicient in amoimt to unite with both the hydrogen and carbon constituents, combines by reason of greater aflm- ity with the hydrogen, leaving the carbon free, but so much raised in temperature that it becomes incandescent, thus furnishing light during the t5itremely slight interval elapsing in its passage to the outermost portion of the flame, where, as has been stated, it is oxidized to carbon dioxide. The same process goes on in the combustion of illuminating gas and oils, the luminosity of the flame being due to the incandescent particles of carbon in the breaking up of the hydrocarbon compounds into their elements. A mixture of gas and air, such as occurs in the use of the Bimsen burner, gives oif little or no light, since each particle of carbon is provided with sufficient oxygen to convert it at once into carbon dioxide, and so incandescence cannot occur. If the air supply to the interior of the flame is shut ofP, luminosity is produced at once. If the area of the outer surflice of an ordinary gas flame is so small that atmospheric oxygen cannot be taken up sufficiently fast to imite with all the carbon arri%ang at the outer part of the flame, the unoxi- dized carbon becomes cooled below the ignition point and is given ofl' in the form of smoke. Defects in the burner or excessive richness in hydrocarbons may cause smoking during combustion, the supplv of air being too small to consume the carbon. The introduction of a cool surface into the luminous portion of the flame causes deposition of soot thereon. If the area of the flame is made too large by turning on a large volume of gas under high pressure, the gas is projected so far that it comes in contact with sufficient atmospheric oxygen to burn a large part of its carbon and hydrogen simultaneously, and, as a result, the excess of gas is consmned without luminosity and wasted. 448 HABITATIONS, SCHOOLS, ETC. Gas Burners. — The best of the burners in most common use is known as the hdfs-wing, from the shape of the flame. The tip is hemispher- ical, and is provided with a single straight slit, through which the gas emerges in a thin flat sheet. Another, known as the Ji--e. The gas, issuing from the holes, forms a circular flame, which is provided with an abundant air supply which passes upward through the perforations of the holder for the chimney, which is an essential part of the apparatus, and through the central hole of the burner as well. The chimney should be of proper diameter and height to insure an air supply adequate for com- plete combustion of the gas. The U'(l.