PURE AND IMPURE FOODS. 



231 



true of the body and of all mechanical 

 devices for performing work. The heat 

 given off from the body shows how much 

 material was burned in it. If mechanical 

 work was performed by the subject, the 

 amount can be measured by suitable de- 

 vices. In some of the experiments the man 

 worked a stationary bicycle which ran a 

 small dynamo. In this and other ways all 

 the energy given off by the body was 

 measured and the total energy produced 

 was compared with that introduced in the 

 food consumed. With the respiration calori- 

 meter it is possible to compare different 

 foods and their capacity for producing 

 work, also to learn the quantities of food 

 required for certain kinds of work. In 

 other words, this apparatus can be used, 

 among other things, to test the value of 

 foods in the human machine just as other 

 devices are used for testing the value of 

 fuel for producing work in an engine. It is 

 a delicate apparatus, as will be seen by the 

 fact that it measures readily the heat pro- 

 duced by the combustion in the body of 

 the small quantities of food necessary to 

 supply the energy used when a man rises 

 from a chair and sits again as slowly as 

 possible. 



The uses of the respiration calorimeter 

 for studying the theories of nutrition are 

 many. It may be and is used for the 

 study of many practical problems. Others 

 are more popular. While some of the fea- 

 tures of this apparatus were suggested by 

 earlier forms devised by European investi- 

 gators, the essential features are original. 

 It is the first device of its kind combining 

 successfully a respiration apparatus and a 

 calorimeter. As an indication of tne way 

 it is regarded by other investigators, it may 

 be mentioned that the German government 

 and the Austrian government have each 

 appropriated considerable sums for the con- 

 struction of similar respiration calorimeters. 



CORDIALS. 



The cordials which are so frequently 

 served after dinner owe their flavor, if genu- 

 ine, to fruits, aromatic herbs, and similar 

 articles. Cordial making is an old prac- 

 tice, having been carried on, for instance, 

 at some European monasteries for genera- 

 tions. Cordials of domestic manufacture 

 are favorites in many families in France 

 and Germany, and old recipes are carefully 

 followed. Home-made cordials are not un- 

 known in America ; the black currant cor- 

 dial, sometimes made in New England, 

 being without doubt a direct descendant 

 of an English ancestor. Cordials fre- 

 quently contain a large percentage of alco- 

 hol. They are usually of pronounced color, 

 which they should own, at least theoreti- 

 cally, to the fruits, flowers, etc., used in 



their manufacture. The home-made cor- 

 dials seldom, if ever, contain added colors. 



The sale of cordials in the United States 

 has grown markedly in recent years. They 

 are liked for their peculiar flavors, but are, 

 without doubt, frequently served on ac- 

 count of their attractive colors. An ex- 

 amination of bright colored cordials was 

 undertaken recently at the Connecticut 

 State Experiment Station at New Haven. 

 According to the investigators, this was 

 done, not because such goods are seri- 

 ously adulterated, but because the cordials 

 illustrate most strikingly the extreme to 

 which the present mania for colored food 

 may be carried. 



Twenty-nine brands were found on sale 

 in the State and analyzed. The investiga- 

 tion revealed the following facts : Cremc 

 de menthe is a cordial usually containing 

 13 to 30 per cent of alcohol, 10 to 40 per 

 cent of sugar, a certain quantity of oil 

 of peppermint and possibly other flavoring 

 matter, and having a vivid green color. Un- 

 colored cremc de menthe has little sale, and 

 it is doubtless true that the popularity of 

 the green product is due quite as much to 

 its color as to its flavor. Cremc de violetic 

 and creme de rose had about the same 

 quantity of alcohol and sugar as the creme 

 de menthe. The former was flavored with 

 a violet-like extract (probably orris) and 

 apparently colored with methyl violet or 

 other coal-tar product. The latter was 

 flavored with rose and colored with vari- 

 ous red dyes. Other cordials, such as 

 cremc de cacao and creme de celeri, belong 

 to the same class with those aire:, y de- 

 scribed. 



All but 2 of the mint cordials anal- 

 yzed were colored with coal-tar dyes, 

 usually malachite green or a closely allied 

 color mixed with a yellow dye. The 2 

 brands which were free from coal-tar dyes 

 contained what appeared to be vegetable 

 colors, probably chlorophyl or leaf-green. 

 This leaf-green is harmless. 



Five samples of cremc de violette were 

 examined. With one exception they were 

 colored with methyl violet, a dye com- 

 monly used in violet ink. Of the samples 

 of creme de rose and rose cordials, 2 con- 

 tained a Bordeaux red or a related dye, 

 one fuchsine (magenta), one a ponceau, 

 and one cochineal. In 2 of the miscellane- 

 ous cordials, a coal-tar orange color (tro- 

 peolin) was detected. The creme de cacao, 

 creme de celeri, cremc de cafe and ratafia 

 de cerese examined contained no added 

 coal-tar color. 



The solid matter in all the samples of 

 creme de menthe and most of the other cor- 

 dials consisted largely, if not entirely, of 

 cane sugar, but in 2 of cremc de violette 

 and one of creme de rose it was in part 

 glucose. 



