We studied a cohort of 333 children in kindergarten to determine the prevalence of seropositivity to Toxocara canis, and to detect and measure chronic health effects that might be attributable to past infection. We found that 23.1% of the children had serologic evidence of infection (antibody titer greater than or equal to 1:32), assayed by means of an enzyme-linked immunosorbent assay. Black children were more frequently infested than white children, as were children of parents who did not graduate from high school. In a subsample of seropositive and seronegative children, we found associations between seropositivity and both pica and puppy ownership; we did not find differences in the symptoms and signs that occur in toxocaral visceral larva migrans nor differences in measures of growth and nutrition. No child had ocular toxocariasis although 31.8% (106) of the children had antibody titers greater than or equal to 1:16. In a population in which approximately 20 to 30% of the children show serologic evidence of Toxocara infestation, care must be taken in differentiating toxocariasis-like ophthalmic lesions, due to the potential for the coincidental occurrence of retinoblastoma in a child who is seropositive for the Toxocara parasite.