The Eskind Biomedical Library has a new website; please visit us there. This website will retire on 08/31/16.


Biographical Sketches

Scientific and Historical Articles

Scientific and Historical Monographs

Feeding Devices

Introduction to Infant Nutrition

Prior to the twentieth century, if a mother could not or would not breast feed her child, the infant faced two formidable risks-the risk of inadequate nutrient intake and that of contaminated milk and water. Wet nursing was a substantial career for some women, and the wet nurse was often seen both as a mother surrogate and as essential to the survival of the child. Although physicians were almost unanimous in their support of breast-feeding, they regularly encountered infants receiving other nourishment. It was not always possible to find a satisfactory wet-nurse and mothers and physicians struggled to create substitute sources of milk for the infant (Still, 1931; Wickes, 1953).

The problem of creating an adequate breast milk substitute occupied much of the attention of certain physicians during the last half of the l9th century (Cone, 1976; Cone, 1979). Experience was their primary guide and, as is often the case, it seemed to lead them in differing directions. although cow's milk was the basis of the most common substitutes, goat's milk was strongly advocated by some practitioners. Some believed in the combination of cereals and milk and all had certain favorite formulas that they recommended to their patients (Wood, 1955).

Toward the end of the 19th century systematic research began to provide reliable information that could be applied to the artificial feeding of infants. Arthur Meigs in Philadelphia, as well as certain European physicians, developed extensive information on the basic composition of milk from various sources (Meigs, 1887; Levinson, (1928). This provided the basis for "percentage feeding", a system popularized by Thomas M. Rotch (1896; 1903; 1907) at Harvard, whereby cows milk was modified to yield a composition similar to that of human milk. From about 1890-1910, this method (sometimes called the "laboratory" or the "American" method) dominated more sophisticated pediatric practices. Nonetheless, it was complicated and required reliable, and often expensive, sources of milk, cream and lactose. Some physicians, such as Jacobi, were not convinced of its efficacy. Moreover, there was lack of understanding about the digestibility of the fat and protein in cow's milk. This led to inappropriate modification of milk in some cases (Holt, 1903; 1925).

A reliable and clean source of milk was of major concern to those who were creating artificial formulas for infant feeding (Cheadle, 1889; Rotch, 1896. Especially in the summer, diarrhea due to contaminated milk was far too common and the result was much higher infant mortality in the summer months. Pasteur's well-known work on bacterial contamination was reported in the 1860s but the adoption of pasteurization of milk was not widespread in the United States until after World War I. To address the problem of cleanliness in the 1890s, certified milk was introduced and soon became available from milk laboratories in the major cities (Waserman, 1972). In Boston, and later in certain other major medical centers, laboratories that could supply modified milk to fit a physician's prescription were established (Rotch, 1907). They operated much like specialized pharmacies, filling very precise prescriptions for the composition of the milk to be fed to infants from well-to-do families. Poor families could neither afford the medical advice nor the expensive products of these laboratories.

By 1920, the advances of nutrition science were ready to be applied to infant feeding. Percentage feeding had fallen out of favor because it was too complicated and there was little evidence that it was greatly superior to simpler systems of milk modification. Moreover, pasteurized milk had become widely available and the incidence of infant diarrhea had greatly diminished. In addition, Otto Heubner and Max Rubner (1899) in Germany had provided experimentally based information on the caloric needs of infants. Their data formed a scientifically based standard for feeding, a standard based on calories rather than the more elaborate percentage feeding schemes.

Scurvy and rickets, however, were epidemic in certain American infant populations, as they had been in parts of Europe for over a century. The American Pediatric Society (founded in 1888) focused on scurvy, creating a system to summarize the current knowledge of cases (Cone, 1976). Later, Alfred Hess (1920; 1929) in New York City was a leader in demonstrating the value of orange juice and cod-liver oil in treating or preventing these diseases. Improved sanitation plus adequate attention to nutritional needs resulted in much lower mortality among those infants who were artificially fed.

Additional advances were made in finding ways to simplify the modification of milk for infants so that mothers or other care-givers could easily make the appropriate adjustments themselves. McKim Marriott in St. Louis was a strong and influential advocate of the use of evaporated milk as the basis from which simple modifications could be made (Marriott and Schoenthal, 1929). Phil Jeans in Iowa City provided leadership within the American Medical Association in providing up-to-date information to physicians regarding nutrient needs of infants and how to supply them.

From the mid-1860s, commercial concerns used the latest information to modify their products in keeping with the existing standards and to assure they were easily employed in infant feeding (Liebig, 1866; Rotch, 1903). At first they were based on very inadequate information and could be actually harmful to the infant. In his textbook, Rotch (1896) stated: "My own opinion in regard to patent foods is that they must necessarily be unreliable." As scientific knowledge improved, however, these products greatly improved and their use in infant feeding was encouraged (Committee, 1965). Nonetheless, researchers continued to struggle to understand the nutrient needs of infants and how best to support infant growth and development (Fomon, 1965; 1967)

By the 1950s attention was drawn to the question of how soon foods in addition to milk should be included in the infant's diet (Committee, 1958). In l900 it was common to recommend additional foods only when the infant was about 12 months old. That recommendation rapidly shifted downward, however, until by the 1960s some infants were receiving cereals and other foods within the first month of life (Guthrie, 1966).

Scientific progress during the Century of the Child has greatly improved the well-being of infants through improved understanding of infant nutritional needs and the basis for those needs (Foman, 1967). Nonetheless, still better understanding is needed.

Patricia Swan, Ph.D.