The most scientific modification of milk which has ever been attempted is that known as the " American or percentage " method. It has been evolved by the genius and enthusiasm of Dr. T. M. Rotch of Harvard. Every precaution is taken which human ingenuity could suggest to render the source of the milk supply as good as possible, and to preserve the milk from contamination during the period between leaving the cow and reaching the infant. As a result of the special method of feeding the cows, the milk has an almost uniform percentage of the solid contents all the year. The process through which the milk passes is a somewhat prolonged one, consisting of separation, recombination according to the physician's prescription, bottling and sterilizing. The modifying materials are kept for use in glass vessels at a temperature of 40° F. to prevent the growth of bacteria. The milk is delivered at the laboratory within a few hours of milking, being conveyed in ice-chests in which the temperature does not rise above 45° F. On arriving at the laboratory the milk is divided by the separator into cream, separated milk, and any dirt or foreign matter which may have found entrance. The cream has a fixed percentage of fat, namely 16 per cent. In making up the milk for an infant in accordance with the physician's prescription, the clerks have the following materials ready in glass vessels; (1) The 16 per cent cream obtained from the separator, (2) the separated milk to be used for securing the proper percentage of proteins, (3) a 20 per cent solution of milk-sugar in distilled water, and (4) lime-water. With these materials the modifying clerk combines each infant's food according to the prescription before him, and then pours it into the glass tubes from which the infant is to be fed. These tubes, each of which contains the milk for one meal, are closed with a plug of non-absorbent wool, and placed in baskets with compartments equal in number to the meals which the infant is to receive daily. The baskets and their contents are placed in the sterilizer, and after having been subjected to the required temperature, are taken to the cooling tank where the temperature of the food is reduced to 28° F. They are then ready to be sent to the customer's house.

As an example of how laboratory milk is ordered by the physician, Dr. Rotch gives the following prescription for a healthy infant of four months.

Per cent.

Reaction slightly alkaline.

Fat.....

4

Number of feedings, 7.

Milk Sugar .

7

Amount at each feeding, 4 1/2 oz.

Protein.....

150

Heated for, 20 minutes. Heated at, 167° F.

More recently by the use of whey, the protein constituents can be split up and ordered as casein and lactalbumin in the percentage desired.

The advantages claimed for this method are that the composition of the milk is exactly determined so that it may be made the same as healthy breast milk, or may be altered by the physician to suit the infant's requirements. The milk is sterile, free from dirt, and has not lost its quality of freshness, unless sterilized at too high a temperature or for too long a time. It is said to be suitable for healthy infants and to prevent digestive disturbances, as well as cure them, in those who have not had a similiar diet.

It is possible that in America the infantile stomach does require some such milk, for one cannot but be struck by the amount of writing and the diversity of opinion in that country on the subject of infant feeding. In this country there are but few milk laboratories, and the need for them does not appear to be very urgent. When pure fresh milk is obtainable the needs of the infant can be met by simple home modifications, and when it is not obtainable no laboratory treatment will render stale milk wholesome. The retail price of laboratory milk, about 8d. per quart, will tend to put it beyond the reach of most. Opinions still differ as to the effects of this mechanical treatment of milk, the splitting it up and the recombination. Some hold that both the proteins and the fats are rendered less digestible. It is admitted by all that a considerable amount of practice is required before the exact percentage amounts necessary for different infants can be learned. Many clinicians find themselves unable to determine those minute percentage differences in the proteins, fats, etc., which are placed at their disposal by this method. Many hold that such minute differences are not called for in infant feeding. In the pre-labora-tory days, the differences of opinion as to milk feeding were ascribed to the want of any definite composition in the milk, now the differences of opinion seem to take place over which standard to select, when the various solids of milk can be differentiated to two decimal points. The tendency in this country, and we think it is a wise one, is to have as little tampering with the milk as possible, and this system of separating, recombining and sterilizing milk opens the door to many possibilities of error in the preparation. No one can tell that the milk supplied is the same as the milk ordered. One ignorant or careless dispenser may be productive of much mischief.

Before recommending the use of laboratory milk we should like to see some more evidence of the necessity for it in this country, and some more unanimity of opinion as to its advantages in America.

Many dairies in this country subject the milk to a process of separation and combination with cream and sugar in order to form a mixture which is known as "humanized milk." It is supposed to resemble breast milk in chemical composition. This preparation has no advantages over milk which has been modified at home, and the composition of the mixture cannot be regarded as uniformly the same.