A longitudinal investigation of the health effects and reservoirs of Glardia was undertaken during 1984–1985 in 40 households located in the rural Nile Delta region of Egypt. Stool specimens obtained once weekly for six months from 2- 4-year-old children were cyst- or trophozolte-positive in 42% of the 724 examined. Only one child remained Glardia-negatlve during the study. The mean duration of excretion in Giardia-posltlve children was seven and one-half weeks with a range of one to 17 weeks. Mucus was present in 52% of all stools collected, and fecel leukocytes were observed with surprising frequency in the absence of identifiable pathogens. Clinical symptoms of illness were frequently observed within a month before or after Glardia excretion in stool of children, but a statistical inference of association was not demonstrated. Seventeen per cent of 697 specimens ob tained from their mothers were Glardia-posltive for a mean duration of four weeks and a range of one to 18 weeks. A total of 962 specimens were collected from 13 species of household livestock. Glardia was detected in 22 specimens from cows, goats, sheep, and one duck. Glardia cysts were detected in three of 899 samples of household drinking water. The ubiquity of the protozoan as well as the failure to show an association between infection and symptomatic illness argue against the administration of Glardia-speclflc drugs to children in settings where the risk of reinfection is high and for whom intestinal insults are both varied and constant.