(1) By mixing into the milk some substance which shall separate the particles of albumen from each other, and so cause it to form into smaller masses. (2) By partial predigestion. To accomplish the first, it is necessary to use some diluting substance of a harmless nature; if it be nutritious, so much the better. For this, Mellin's food, barley-water, veal broth, lime-water, and gelatin are recommended. Mellin's food is a partially predigested grain, in such a condition that it can be assimilated by the infant; barley-water is valuable for its potash salts, in which cow's milk is deficient, and which the growing babe needs; veal broth is rich in lime; and lime-water neutralizes the acid of the gastric juice, so that milk is not acted upon so strongly, and consequently forms into a lighter curd. The second method is that of partial predigestion, and is accomplished by the use of peptonizing agents, among which Fairchild's peptogenic milk-powder is good (directions for its use will be given later). On account of the expense of these preparations it is not probable that they will come into general use, except in cases of sickness. It is therefore evident that dependence must be placed almost entirely upon attenuants to render the casein of cow's milk more easily digestible. Probably for this Mellin's food is as good, if not better, than any other of the recommended preparations. It is not injurious, is nutritious in itself, and is a good diluting agent, causing milk to form into looser curds than it would otherwise do, and it contains sufficient sugar to require no further addition of this substance. Now arises the question whether milk shall be sterilized for infants' feeding. The weight of evidence seems to be as follows: if it is possible to see the conditions under which the cows live, and to _know_ that they are unquestionably good, that the animals are in perfect health, that the milk is drawn from cleansed udders into cleansed vessels by clean hands, kept in a cool place, and used fresh, then it is probably wise not to sterilize it. All milk otherwise obtained should be made sterile before using, and as soon as possible after milking. Looking to the standard--human milk--there are no organisms in it. That alone is sufficient reason why cow's milk should be freed from them.[55] Again, most bottle-fed children do well during the cold weather of autumn and winter; in summer the mortality is very great among them, especially in the poorer districts of large cities. It is well known that the chances for life with children nourished by mother's milk are greater than with those artificially fed. Why should this be? There is no doubt that it is owing to the presence in cow's milk of extraneous substances, the products of bacterial growth--products which are often absolute poisons; and it is highly probable that cholera infantum, in a vast majority of cases, may be traced to the action of such poisons.