There are three classes of infants who require weak-milk mixtures to begin with: namely, the baby who has been previously nursed and whose mother's milk has utterly failed; the baby just weaned; and the infant whose power to digest is low. If these children were six months old, and the formula best suited to them is unknown, we must begin with a formula suited to a two- or three-month-old child and quickly work up to the six-month formula, which may often be accomplished within two or three days. THE BOTTLE-FED BABY When a baby is getting on well with his food, he should show the following characteristics: He should have a good appetite; should have no vomiting or gas; he should cry but little; and he should sleep quietly and restfully. His bowels should move once or twice in twenty-four hours. His stool should be a pasty homogeneous mass. He should possess a clear skin and good color. He should show some gain each week--from four to eight ounces--and he should also show mental development. As long as a baby appears happy and gains from four to eight ounces a week and seems comfortable and well satisfied, the feeding mixture should not be changed or increased. MAKE CHANGES GRADUALLY In our experience with the artificial feeding of infants, we have come to look upon the practice of gradually changing the food formula as the most important element in successful baby feeding. We recall one mother in the suburbs who came to us with her baby who had been feeding on a certain proprietary food. She declared that it "just couldn't take cow's milk." She admitted "it was not doing well," and so she would like to have help. The baby was old enough, had it been normal, to have been taking whole milk for some time. We recall our having the mother prepare the proprietary food just as she had been used to preparing it, and each day we had her throw away one-half ounce and put in one-half ounce of whole milk, this mixture she fed the baby for two days. The next time, we had her take out one ounce of the mixture and put in one ounce of whole milk, which we fed the baby for three successive days; and then one and one-half ounces were substituted which was fed to the baby for four days; and thus we carefully, slowly, and gradually withdrew the proprietary food and substituted fresh, certified cow's milk. It took us a month to complete the change, but we are glad to add that it was done without in the least disturbing the child. Now, had the change been made abruptly--in a day or two, or three days--the baby would probably have been completely upset, while both the mother and the doctor would have been greatly discouraged. Many mothers and even some physicians have jumped from one baby food to another baby food; they have tried this and they have tried that, until the poor child, having been the victim of a number of such dietetic experiments, finally succumbed.