Most density-pathology research has assumed that dense living conditions have ‘ pathological’ effects on urban dwellers. This study considers the possibility that both low and high density may be problematic and that in-unit and out-unit density may have interacting effects. The data included all first admissions from Chicago to 44 in-patient psychiatric facilities in Illinois during 1960–61. After controlling for socioeconomic status (SES) and race, living alone, which defines low in-unit density, was associated with high admission rates, but only when the outside density was low. Among those not living alone, the contribution of density to admissions was agelinked, with interactions between in–unit and out-unit density observed in the older ages. The results lead to the general conclusions that the negative model guiding earlier research is inadequate and that in-unit and out-unit density have interacting effects on the risk of psychopathology.