Szmuness, W. and A. M. Prince (The New York Blood Center, 310 E. 67th St., New York, N.Y. 10021). The epidemiology of serum hepatitis (SH) infections: A controlled study in two closed institutions. Amer J Epidem 94:585–595, 1971.—In order to test some hypotheses suggested by a pilot study, 3 independent samples of pairs of institutionalized mentally retarded subjects, selected by means of a computer program, were tested for the presence of the SH antigen. Examination of 196 mongols revealed them to have 2.5 as high a frequency of SH carriers as strictly matched non-mongols (point prevalence 20.4% and 7.7%, respectively). The higher prevalence in mongols appears to be primarily associated with a longer persistence of antigenemia. The chronic carrier state, in both mongols and non-mongols, is most probably not lifelong. A study of 221 pairs consisting of patients admitted for the first time to institutions for the mentally retarded under 5 years of age, and matched controls admitted at age 10 years and more, has confirmed the importance of early age of exposure to infection as a risk factor. Patients admitted at an early age had a 3.4 fold greater risk of becoming chronic carriers. Employment of strict matching procedures significantly reduced the previously reported dependence of prevalence upon sex. Regardless of diagnosis, patients having the 3 most prominent risk factors were found to carry the antigen 5–6 times more frequently than patients without these risk factors. Patients with a past history of overt viral hepatitis tended to be chronic carriers more frequently than those without such a history. Indirect evidence suggested that the chronic carrier state usually originates from anicteric infections. Many lines of circumstantial evidence support the hypothesis that serum hepatitis virus is transmitted by nonparenteral routes.