This is the addition of a small quantity of a "latent colour" to the margarine, so that, although it appears yellow, like butter, its colour can be changed by the application of a single reagent to pink or blue, and its presence thus revealed in a mixture of butter and margarine. Several years ago an attempt was made in some of the United States to compel manufacturers of margarine to colour it pink, so that it could not possibly be palmed off as butter, but as this law was found to have the effect of stopping the sale of margarine altogether, it is no longer enforced. Various substances have been suggested as suitable for the latent colouring matter, such as starch, which turns blue on contact with iodine, and certain colourless coal-tar derivatives which change to pink upon the addition of an alkali or acid. There are numerous objections to the use of some of these compounds. Thus, starch may be washed out of the margarine by a simple treatment with water, while a coal tar derivative that turns pink on contact with an alkali is too sensitive an ingredient for everyday use. A far more satisfactory substance than any of these was found in the oil derived from sesame seed. This is a wholesome oil with a fragrant odour and pleasant taste, which is largely used as a salad oil in certain parts of Europe. It is one of the few vegetable oils that can be detected by means of a special colour reaction; for on treating the oil with a particular reagent it gives a bright rose colour, and the test is so sensitive that it will detect the presence of even a small percentage of sesame oil in other fats. A compulsory addition of a small amount of sesame oil to all margarine, therefore, affords an absolutely certain means of recognising the margarine subsequently. The first country to adopt this plan was Germany, where a few years ago a regulation was made that all makers of margarine must use 10 per cent. of sesame oil with the other ingredients. Belgium has also adopted the same plan of earmarking the margarine produced in the country, and has thus simplified in one direction the problem of detecting petty adulteration. A similar problem has to be solved in dealing with milk, the proportion of cream in which varies naturally to such an extent that it is possible to add a considerable amount of water to a rich milk without bringing it below the level of a poor but genuine milk. When such a sample of milk has been condemned, the analyst has often been confronted by an appeal to the cow herself. But even the specious notice which was for years to be seen over a dairy: "Our customers may come and see the cows milked into their own jugs," is no proof that the fluid they yield is necessarily "milk."