How to distinguish a good article of food from a bad one, when both are in their natural state, is within the province of the cook-book. In this place will be pointed out only the adulterations of food, and those methods of detecting them which can be used by householders who have no special knowledge of the instrumental and chemical means which are generally necessary.

Adulterations are of two kinds: those which injure the consumer, and those which simply cheat him. The following details are chiefly taken from the New York State Board of Health report for 1881-’82, the name of the analyst being in each case appended.

Arrowroot.—Often mixed with cheaper starches. Twenty-three samples examined: seventeen were arrowroot, one was arrowroot and tapioca, two arrowroot, tapioca, and potato, and three tapioca and potato. Harmless. (E. G. Love, Ph. D.)

Bakers’ chemicals.—(Saleratus.) This was originally bicarbonate of potash, but the name is now applied to the bicarbonate of soda. Twenty samples: none adulterated.—(Baking-soda.) Twenty-three samples: twenty samples unadulterated. One contained 25 per cent of gypsum; one same quantity of gypsum and a little starch; one a large amount of sulphate of soda and 17 per cent of carbonate of lime.—(Cream of tartar.) This is the bitartrate of potash. Twenty-seven samples: sixteen adulterated and in some not a particle of cream of tartar found. Six adulterated with terra alba (gypsum) and starch, one with starch alone, two with starch, terra alba, and acid phosphate of lime. Six had tartaric acid and no cream of tartar. In eight the amount of terra alba was found to vary from 3.27 to 93 per cent. Five samples contained over 70 per cent of this injurious adulteration. (Love.)

Baking-powders.—Contain bicarbonate of soda and some acid or acid salt, which combine when water is added and evolve carbonic-acid gas. There are four classes in use. One contains cream of tartar, one tartaric acid, one the acid phosphate of lime, and one potash or ammonia alum. Many powders contain a salt of ammonia. The pungent odor of this substance prevents its use in any but the smallest quantities, and it can not affect the wholesomeness of the powder. Flour or starch forms an ingredient of many powders to prevent a premature combination of its constituents and a consequent deterioration of the powder. Eighty-four samples: seventy-three had flour or starch; thirty-five contained ammonia. Eight adulterated with terra alba, phosphate of lime, or tartrate of lime. As a rule, these powders are harmless. (Love.)

(Alum.) There does not seem sufficient evidence as to the injurious effects of alum upon the human system to warrant legislation against it. (Love.)

Beer.—Most adulterations are harmless. Corn, rice, wheat, glucose, starch, potatoes, etc., are used in making beer, and in many countries are allowed by special laws (as in England and Germany). The use of substitutes for hops within the last few years is hardly possible to believe, since hops have been so cheap, in fact, almost the cheapest bitter, and are, moreover, as every brewer knows, the best material to preserve his beer. (F. E. Engelhardt, Ph. D.)

Brandy.—Cognac brandy is naturally colorless, but public taste demands a brown color, which is imparted by a mixture of caramel (burned sugar). A very large proportion of the brandy in the world is made of corn-spirit colored with burned sugar and flavored with oil of cognac (an essential oil derived from the lees of wine; twenty-five hundred pounds of lees make one pound of oil.) Here is one recipe: “To every ten gallons of pure spirits add two quarts New England rum or one quart Jamaica rum, and from thirty to forty drops of oil of cognac cut in one half pint of alcohol; color with burned sugar.” Twenty-five samples examined: sixteen contained fusel-oil, six had traces of it, and three none. (Engelhardt). The only injurious ingredient (besides alcohol) in the artificial brandies seems to be the fusel-oil contained in the corn-spirit or whisky used as a base. Rub suspected brandy on the palm of the hand until it has evaporated. Fusel-oil may then still be detected by the smell. Good brandy will leave no odor.

Bread.—Ten samples: no adulteration. (Love.) In 1873, Elwyn Waller, for the New York City Board of Health, examined fifty-one samples, of which forty-one were unadulterated, and ten contained traces of copper or alum.

Butter.—May be mixed with oleomargarine, and the adulteration is hard to detect. Oleomargarine is more crumbly than butter in cold weather. Often colored with annotto, and, as this sometimes contains a little sulphate of copper, a trace of copper may occasionally be found in butter. As a rule, adulterations of butter are harmless.

Candy.—“Taffy” and gum-drops are almost all glucose. Coloring-matters usually harmless, but of ten samples of yellow candy, seven contained chromate of lead. (W. H. Pitt, M. D.) Candy often contains terra alba, flour, and gum-arabic. The only injurious ingredients usually found are terra alba (recognized by its grittiness and insolubility) and the chromate of lead. On account of the latter it is best to avoid yellow, green, and orange candies.

Canned fruits and vegetables.—Eighteen samples, including peaches, plums, grapes, strawberries, cherries, blackberries, olives, mushrooms, corn, beans, succotash, tomatoes, pumpkin, and peas. No adulteration found. Attention was given to the possibility of the chemical reaction of the fruit acids upon the inner surface of the cans, whereby salts of tin and lead might be produced, rendering the contents in some degree poisonous. There was no evidence of their presence. Some of the articles were canned over a year before. (S. A. Lattimore, Ph. D.) Per contra, other analysts have found from .1 to 2.3 grains of tin to the can. There is no evidence, however, that this amount of tin in solution is injurious, and the recent investigations of Hall seem to show that the fruit acids do not act appreciably on the tin or the lead of the solder, so long as the air is excluded. Cans once opened should therefore be emptied, and not left partly full.

Canned meats.—No adulterations and no tin or lead found. The heads of all cans should be slightly concave. This shows that the contents were hot when the can was sealed. If the heads are convex, the contents are decomposing.

Cereals.—Ninety-four samples, including wheat and Graham flour, farina, oatmeal, rye, barley, corn-meal, rice, buckwheat, sago, and tapioca. Two adulterated. Adulterations consisted in mixture of other cereals, and were harmless. (Love.)

Cheese.—Skim-milk cheese often contains lard, put in to replace the butter which has been removed. This adulteration improves the quality of the cheese and often can not be detected by experts. One sample of cheese, which had caused sickness in those who ate it, was examined, and, although it also made the analyst ill, no known poisonous substance could be detected in it. (G. C. Caldwell, Ph. D.) All yellow cheese is colored with annotto, which can therefore hardly be considered an adulteration. The rind is sometimes washed with a mercurial or arsenical solution, to protect it from insects, and should therefore never be eaten.

Cocoa and chocolate.—Six samples. None adulterated. (Lattimore.)

Coffee.—Thirty-five samples of unroasted, three of roasted unground, twenty-one of ground, three of coffee extract. Of the thirty-five unroasted, in five a few grains were found which had been slightly colored or faced, apparently with Prussian blue; the three roasted unground were pure; of the twenty-one ground, nineteen contained chiccory, beans, wheat, rye, etc. One sample consisted entirely of roasted hominy. Three samples of coffee extract consisted chiefly of caramel and licorice, and contained no coffee. (Lattimore.) Pure coffee swims in water, and colors it slowly. Chiccory sinks and colors water rapidly. Peas sink and color water slowly. Rye sinks more rapidly than coffee, and colors water more quickly. Ground coffee is hard and crumbles between the teeth; chiccory is soft and does not crumble.

Gin.—No injurious adulterations detected. Twenty-five samples. (Engelhardt.)

Honey.—Three samples: two pure. One, labeled “white-clover honey,” contained 50 per cent of artificial glucose. The presence of added glucose is indicated by the turbidity produced by oxalic acid in a solution of the honey in distilled water. This turbidity is due to the presence of gypsum in artificial glucose—a substance which is not contained in pure honey. (Pitt.)

Horse-radish.—Often harmlessly adulterated with grated turnip.

Isinglass.—Two samples. Both were common gelatine. (Chester.)

Jellies.—Fruit-jellies are often simple apple-jelly, flavored with artificial essences and colored with aniline. Safest not to use them unless their source is known to be trustworthy.

Lard.—Twenty-eight samples. Fifteen pure; the rest contained water. Good lard should melt to a clear fat without sputtering.

Meat.—Sound fresh meat is pale red when first cut, the surface after exposure turning to a deep red. The meat of animals that have died a natural death is of a deep purple color, not having been bled. The greatest danger in meat, however, is that due to the presence of trichinæ, which are killed by a temperature of 160° Fahr. All forms of pork should therefore be cooked thoroughly before eating. (Chester.) Poisoning (vomiting, cramps, and diarrhœa) produced by meat is generally caused by some kind of preserved meat (sausages, pickled meats, etc.), and is probably due to the presence of fungi.

Milk.—Frauds consist generally in adding water or removing fat (skimming). As the milk of healthy cows varies in composition within certain limits, it is necessary to have a standard of purity, which has been fixed upon in New York as follows: Nearly 1,000 cows have been examined, with reference to the specific gravity of their milk, in New York, New Jersey, and Connecticut. The maximum specific gravity was 1.039 in milk of an Alderney cow. The minimum for normal milk from a healthy cow was 1.029. The specific gravity is determined by an instrument called a lactometer, on which 0 stands for 1,000, the specific gravity of water, and 100 for 1,029, that of the poorest milk from a healthy cow. The composition of such milk, adopted by the English Society of Public Analysts and the New York City Board of Health, as a result of fifty analyses, is as follows:

Method of using the Lactometer.—Put the milk in a vessel so deep that the lactometer, when introduced and allowed to float, shall not touch the bottom. Notice the reading of the scale at the surface of the milk. If it is less than 100, it gives the percentage of milk in the sample. For example, if the reading be 80, the sample contains 80 per cent of milk and 20 per cent of water.

Sources of Error.—Milk very rich in cream may possibly, though not probably, register less than 100, but its very appearance will show that it has not been thinned by water or by skimming.

Skimmed milk, especially if a little salt has been added, may register high above 100, but its thinness and blueness will show that it has been doctored.

Condensed milk was carefully analyzed, and found to be unobjectionable. (C. E. Munsell, Ph. D.)

Olive-Oil.—Often adulterated with poppy, cotton-seed, ground or peanut, sesame, rape-seed, colza or beechnut oil, all harmless. Sixteen samples; nine adulterated. (Caldwell.)

Pickles.—Nine samples. None contained copper or any other metal. The only sample that possessed a suspiciously green appearance was found to contain alum. (Lattimore.)

Rum.—Twenty-five samples. No objectionable additions found. (Engelhardt.)

Sirups.—Three samples of maple-sirup. Two were pure, and one, manufactured in Chicago and sold in cans, contained 35 per cent of artificial glucose. In 1870 Dr. Chandler found .02 per cent of tin in each of two samples of sugar-house sirups. This represents .8 grain of tin to the gallon. A common adulterant of sirups is glucose, which diminishes their sweetening power, but is not considered injurious.

Spices.—One hundred and eighty samples, comprising mustard, ginger, allspice, cinnamon, cassia, cloves, white, black, and red pepper, mace, and nutmeg. One hundred and twelve were adulterated from 40 to 81.8 per cent. All the adulterations were harmless (wheat and buckwheat bran, hulls of different seeds, middlings of corn-meal, stale ship’s bread, peas, beans, etc.). No poisonous substance was found. (Lattimore.)

Sugar.—One hundred and sixteen samples, principally collected in New York city. Care was taken to secure the samples from different sections of the city and from all classes of stores. Of these, thirty-four were microscopically clean, fifty-four slightly contaminated with dust, twenty-two contained considerable dirt, and six were very dirty. But in no case was there an intentional addition of insoluble mineral matters. Of forty-nine white sugars, all were pure; of sixty-seven brown sugars, four were adulterated with glucose. (A. L. Colby, Ph. B.)

There have been many exaggerated statements put forth regarding the adulteration of sugar. In 1870 Dr. Chandler reported to the New York City Board of Health that sixty samples of sugar bought at small groceries were found pure and unadulterated without exception. In 1872, Elwyn Waller, for the same board, examined one hundred and nine samples of powdered sugar, but found no adulteration.

Powdered sugar is quite generally believed to be adulterated with gypsum or flour. As both of these adulterants are insoluble in water, it is easy for any one to convince himself of the purity of sugar by dissolving it in water.

Teas.—Forty-three samples of green tea, and eighteen of black. Many were cheap and of very inferior quality, some mere tea rubbish, yet no leaf, or fragment of a leaf, which was examined, could be considered anything but tea. No adulterations were found, and even the admixture of exhausted leaves could not be positively asserted. (Lattimore.) Suspected leaves should be wet and spread out, and then compared with leaves known to be genuine. It is said that exhausted leaves of green tea are often colored or “faced” with plumbago, Prussian blue, soapstone, etc., so artfully that only an expert can detect the fraud. Black tea is generally pure.

Vinegar.—Four samples, all poor, but not adulterated, unless with water. (Lattimore.)

Whisky.—Twenty-five samples. Fusel-oil decided in twenty, and traces in the rest. No injurious adulteration found. “It is evident that the addition of water and coloring-matter is practiced more than any other adulteration.” (Engelhardt.)

Wine.—“A good wine should be transparent, and should have a bouquet. When pouring it into a glass, it should sparkle. A sour taste is always a sign of poor wine. Dizziness and headache are not produced by drinking pure wine. Cloudy, discolored, highly colored wines are suspicious.” There are various substances used in the manufacture of wine which should be classed as adulterations, e.g., calcined plaster is added to the grape-juice during fermentation (so-called plastering); in this way is formed an insoluble tartrate of lime, and a soluble sulphate of potash, the latter having a bitter taste and acting as a purgative even in small doses. The French Government forbids the sale of wine containing over 0.2 per cent of sulphate of potash. This process also leads to the formation of acid sulphates and free sulphuric acid in wines. Plastering of wines is practiced in Spain, Portugal, and the south of France.

Wines are often fortified by the addition of brandy, cologne spirit, or French spirit, to arrest fermentation. Ports and sherries are almost invariably so.

Red wines are often colored with logwood, Brazil-wood, fuchsine, cochineal, black hollyhock and red poppy flowers, alkana-root, red beets, cherries, whortleberries, elderberries, pokeberries, etc. It is very difficult to detect these, and fuchsin is the only one that is poisonous. Carpené gives the following very simple method to decide whether a red wine is naturally or artificially colored: Take a piece of good, white burned lime, break it into two pieces, smooth the surfaces by a knife or file, and place a few drops in succession on the same spot of the smooth surface, and observe after a few minutes the color produced. Natural red wines give a yellowish-brown spot; colored with fuchsine, or Brazil-wood, a rose-colored spot; colored with logwood, a dark-violet spot; colored with cochineal, a reddish-violet spot; colored with black hollyhock, a yellowish-brown spot; colored with pokeberries, a yellowish somewhat red spot. (Engelhardt.)

Another test is to concentrate the wine, and dip in a piece of pure white woolen-yarn. The natural red coloring-matter of wine does not dye without a mordant, while fuchsine and cochineal dye it red or pink.