The effect of the nutritional regimen on infection was tested by inoculating the animals with either one of four pathogens (Mycobacterium tuberculosis var. bovis, Mycobacterium fortuitum, Staphylococcus aureus, Klebsiella pneumoniae type C), and by observing the survival time. The infective dose was administered by either one of three routes: intravenous, intraperitoneal, or air-borne (aerosol).

In some experiments, the animals were maintained in groups of five throughout the tests. In other experiments they were housed in individual cages. This difference in housing did not affect the results in a detectable manner.

Mice fed diets containing 5 or 8 per cent casein as sole source of amino acid (except for cystine supplementation) proved more susceptible to the experimental diseases than did mice fed diets containing 15 or 20 per cent of the same protein.

Susceptibility to infection developed when wheat gluten, or soybean -protein, was substituted for casein—even in high concentrations (15 or 20 per cent).

In one experiment, mice were fed a diet containing as sole source of amino acids a mixture of soybean and rice flour, so designed as to provide a protein concentration of 15 per cent, with an amino acid pattern similar to that of casein. These animals gained weight at the same rate as those fed a diet containing 15 per cent casein and they exhibited a satisfactory level of resistance to bacterial infection.

The infection-enhancing effect of low casein concentration (5 and 8 per cent) could be corrected by supplementing the diet with the proper mixture of amino acids. This could be done using either synthetic or natural amino acids. In contrast, susceptibility to infection developed when low casein diets were supplemented with unbalanced mixtures of amino acids.

The infection-enhancing effect of gluten diets could not be corrected by supplementing the latter with lysine even though this supplementation markedly improved weight gains in uninfected animals.

It appears in conclusion that the relative proportion of the various amino acids in the diet is as important a factor as their total amount in conditioning resistance to bacterial infections. This effect of nutrition on resistance can be detected irrespective of the route of infection: intravenous, intraperitoneal, or air-borne. Moreover, the effect has been observed with two strains of mice differing markedly in their natural resistance to bacterial infection.